1. Introduction

This paper extends the work presented at the 2018 IEEE International Conference on Big Data [1]

Since Krizhevsky et al. [2] introduced specifically designed CNN (Convolutional Neural Network) architectures, there have been various methods for increasing the object classification rate. [3]-[8] have shown performance to be increased compared to the shallow learning in neural networks. These days CNNs have around a 90 percent object classification rate for unblurred or slightly blurred images. Moreover, algorithms in [9], [10] introduced much faster detection times, an increased number of classes, and object segmentations in addition to the Softmax and linear regression algorithms.

Recently, to reduce misdetections and detection failures on CNNs, research has been conducted, such as on generative adversarial networks (GAN) [11]-[13], GAN with reinforcement learning [14], and Capsule networks [15]. GAN solved many of the problems from adversarial noises. Sara Sabour et al. in [15] have attained considerably better results than CNNs on MultiMNIST with smaller sized training data sets. However, Capsule networks, which are groups of neurons learning to detect a particular object within a region of the image [15] do not perform as well as CNNs on larger images even though Capsule networks improve CNNs’ weaknesses caused by pose information (such as precise object position, rotation, size, and so on). Also, Capsule networks require much more computing resources than CNNs.

Even though these studies demonstrate that there have been many improvements in CNNs, there are still detection failures caused by blurred images or low-resolution images. For example, in [16], 100 images were randomly chosen, and these images were preprocessed at a resolution, which was three times lower than the original images in order to generate cropped or warped images, which are called regions of interest (ROI). Following this, they were interpolated using bilinear or bicubic interpolation and then were tested in [4]. These interpolation methods may cause aliasing effects on the images and make larger regions of interest. Therefore, a new method might be needed to generate less aliased and higher-resolution ROIs from original images in CNNs.

Since CNNs were introduced, the most common image size has been approximately 256×256 pixels. Examples in the Spatial Pyramid Pooling network, Faster R-CNN, and ConvNet are tested for the adequateness of the image size.  Also, in [17]-[20], it was shown that a recurrent neural network model is able to extract features from an image by selecting a sequence of regions and processing each region at high resolution. Therefore, object detections from ROIs are very common in CNNs.

In [2], Krizhevsky et al. described ImageNet, which has 15 million labeled images with 22,000 categories and variable resolution image sizes and adopted down-scaled rectangular images from ImageNet dataset. They reported results on 10,184 categories and 8.9 million images.  [3], [6] classified 21 object classes from ImageNet, PASCAL VOC 2007 [16]. However, most research is conducted with VOC2007 20+1 classes even though there are datasets with more than several hundred categories. Recently [9] trained and tested using 80 object classes in the fastest object detection speed. In future research, more than 200,000 object categories should be considered to distinguish objects such as human beings.

There are a number of practical applications for processing low-quality images. These include surveillance cameras, black-box systems in vehicles, and cameras in mobile phones. However, it may be difficult to increase the quality of images in surveillance camera systems because of the limited capacity for storage, insufficient night vision, and dark image sensors. In particular, current night visioning algorithms do not have procedures for good image quality. In the case of black-box systems in vehicles, vibrations and low electric power consumption may cause poor image compression, inability to zoom quickly, or lack of focus. Finally, in mobile phone cameras, the limits of the zoom algorithm can cause images to be blurry or target objects to be small.

In this paper, we will present our research on improving rates of object classification through preprocessing before classification neural networks.

2. Related Research

Compared to the previous CNN algorithms, [3], [21] have shown more new algorithms for feature extracting or new initial parameters. Paper [3] also improved CNN using “region proposals” technique and sliding window. Their new methods increase the speed of object detection, which allows the images to be processed in almost real-time. Using region proposals, the Fast YOLO model in [9] processed 155 frames per second.

In [6], Keiming He et al. had used variable size image datasets and were able to get the best results for training and testing of images after achieving when the shorter side of the input image was maximum 392 pixels. Also, their results indicated that scale is an important factor in recognition processing. Thus, they suggested that the spatial pyramid pooling model and they found that the detection of objects had better performance among several scaled datasets. The main reason for this was because the objects which were detected often occupied significant regions of the whole image. But cropped or warped images caused to get lower accuracy rates than when using the same model on the undistorted full images.

Upon reading this related research, we were motivated to achieve better object recognition performance in (1) regions of interest and (2) high-resolution regions of interest which are cropped areas from input images. High-resolution image cropping is a solution which can come from super-resolution methods.

Figure 1: Cropping and warping. This image is from [6]

In [22], super-resolution is said to construct high-resolution images by adding of new pixels which have high frequency components to the neighbor pixels, instead of adding simply computated new pixels. The frequency components are from multiple observed low-resolution images or from a single low-resolution image.

There are four categories in super-resolution algorithms such as prediction, edge-based, statistical, and example-based methods. Among them, the space-coding based method, which is a kind of example-based method, is popular nowadays.

Let  denote the high-resolution image desired and  be the kth low-resolution observation. Assume the image system captures iteratively k low-resolution images of , where the low-resolution iterations are related with the high-resolution image  by

where  represents the geometric warp (motion information) on the X,  is the linear space-variant blurring effects,  is the down-sampling decimation operator, and  is the additive zero-mean Gaussian noise. Equation (1) can be represented in matrices or equivalently, .

The obtained model (2) is a classic restoration, and the Expectation-Maximization (EM) estimator or the MAP can be applied in order to restore the image X. Similarly to the above image observation model, statistical approaches in [23] relate the super-resolution restoration steps in iteration toward optimal restoration. Therefore, the super-resolution reconstruction is cast into a full Bayesian framework, and by sum rule and Bayes’ theorem. Finally, if we can estimate W, then X can be obtained as a popular maximum a posteriori (MAP), or X can even be reduced to the simplest maximum likelihood (ML) estimator, which can be treated by an Expectation-Maximization (EM) algorithm in machine learning.

In [24], two kinds of super-resolution algorithms are introduced. Compared with the multiple-image super-resolution algorithm, the single image super-resolution algorithm uses a training step for the relationship between a set of high-resolution images and their low-resolution equivalents. This relationship is then used to predict the missing high-resolution components of the input images. There are three steps in this process: the registration (motion estimation), the restoration, and the interpolation to construct a high-resolution image from a low-resolution image or multiple low-resolution images [25].

In addition to the EM algorithm for single image super-resolution, [26]-[25] present methods that employ a mixture of experts (MoE) to jointly learn the features by separating global and local models (space partitioning and local regressing). These are solved by an expectation and maximization (EM) for joint learning of space partitioning and local regressing. Generative Adversarial Networks (GAN) method in [29] is proposed using generative networks and discriminative networks to restore most likely original images with reducing of the mean squared reconstruction error.

In this paper, we will apply a single image super-resolution method to get better detection rates from a single image dataset.

3. Image Resolution Improvements

There are two approaches to improving input image quality. The first is to apply the super-resolution directly to the input images then have the CNN take the super-resolution image as the input image. The second is to apply the super-resolution only to bounding boxes. The second method may reduce the necessary computing power.

Shaoqing Ren et al. [4] implemented the interpolation algorithm for two purposes. One was to fix the input image size from variable input images, and the other was for the selective search algorithm to extract feature maps from input images.  Detector extends networks for windowed detection by a list of crops or selective search (or EdgeBoxes) feature maps and then interpolates them to fix the size into one of the bounding boxes. Thus, this method may have more effects on bounding boxes for small objects, as shown in Figure 2. In [6], this method was described in detail in Section 4. In future research, we will implement CNN with extended images which are preprocessed using the super-resolution method instead of interpolation. In this paper, we propose using super-resolution as pre-processing before CNN to increase the image size to approximately 492×324 pixels.

Figure 2: Images with either full-size objects or small-sized objects. These examples are from [30]

Thus, we will first describe a single image super-resolution method to enhance the quality of the image in preprocessing before forwarding image data to the input layer of CNN. Then, it is followed by the CNN algorithm to classify preprocessed images.

3.1.  Super-resolution as Pre-processing

In image analysis, there have been distinct improvements related to convolution neural networks. However, we will focus on preprocessing image samples by super-resolution methods. For example, if a cropped 166×110 pixel image is extracted from a larger image of 640×480 pixels, the cropped image is usually too small to feed into the input layer of CNN models to get good results in object recognition. Thus, we will import a super-resolution method before the input layer of CNN or for bounding boxes.

We propose a Mixture of Experts (MoE) model to solve problems in anchor-based local learning, which are optimizing the partition of feature space and reducing the number of anchor points. The MoE model will be solved by Expectation-Maximization (EM) algorithm. The objective of a MoE model is to partition a large complex set of data into smaller subsets via the gating function. This preprocessing is built with the lazy neighbor embedding, anchor-based local learning approach, sparse coding, and deep convolution neural networks, respectively.

Component regressors, , and an anchor point,  have the following relationship as an expert which means that each of the model component classifiers or regressors is highly trained:where  is a low resolution path,  is the corresponding high resolution patch,  is the nearest anchor point for  and  is a continuous scalar value which represents the degree of membership of .

However, we propose a mixture of experts, which is one of the conditional combined mixture models in [31][32], in order to partition an image data into smaller subsets (as gating networks to determine which components are dominant in the region) and to determine the best (as experts to make predictions in their own regions).

In general, this a mixture of experts model can be trained efficiently by maximum likelihood using the EM algorithm in the M step. For every iteration, the posterior probabilities are calculated and reweighted for patches, and then we get LR/HR patch pairs  through the expectation of the log-likelihood as an E-step. During M-step, anchor points and regressors are updated, which is a softmax regression problem. After training, which is saving the trained anchor points and the projection matrices, a super-resolution image from the given input low-resolution image can be constructed by collecting all the patches from regressed low-resolution patches and averaging the overlapped pixels.

In order to compare the performance of our proposed method with that of interpolation methods, in addition to images produced by the super-resolution method, bilinear and bicubic interpolated images will also be generated.

3.2.  Object classification

As the object classification model, we decided to implement the CNN mostly based on the Faster R-CNN [4] because these networks support variable image sizes as the input data of CNN and sliding widow proposal scanning for the convolution network. Most importantly, the detection speed to predict an object is almost as fast as real-time. In YOLO or SSD, a group of an object can be detected, and the Faster R-CNN is enough to meet our goal. As mentioned earlier, Faster R-CNN uses the region proposal approach to detect an object from variable input image sizes. But the size ratio of the region proposal to the whole image is critical to detect the object. As shown in Figure 2, sometimes objects from small bounding boxes are ignored because the object may be deformed or aliased during cropping or warping. With a better quality image, CNN will make better quality cropped or warped bounding boxes. This means we can distinguish our proposed method from other interpolated data.

As the pre-processing, the super-resolution is implemented on MATLAB on Linux and Intel® Xeon. Our proposed neural networks are implemented in two ways. For one example, CNN is implemented only on Intel® Xeon CPU 2.30GHz, and as another example, CNN is implemented based on Intel® Xeon CPU 2.30GHz and eight GPUs in four NVIDIA Tesla K80 GPU boards. Since our purpose is not to find how fast objects are detected but rather how many objects are detected, we do not discuss these different platforms anymore.

Our proposed CNN has built to predict region proposals in terms of region proposal networks (RPN), which are implemented by deep convolution networks (a kind of fully convolutional network  [6]) [2][8]. This RPNs can predict region proposals with wide ranges of input image sizes and aspect ratios in contrast to conventional neural networks. The RPN also runs on pyramids of regression references (or pyramid of anchors), not on pyramids of images. It has benefits on running speeds.

The detection modules are organized as below: Multiple convolution layers convolve an input image and output feature maps. Then feature maps are given to the RPN. The RPN, in terms of a sliding window with a Multi-boxing method, generates a set of rectangular object proposals as a deep, fully convolutional network. A wide range of input image scales is fixed into given anchor sizes. After the RPN, conventional CNN predicts objects using the classification layer and bounds a box around the detected object using the regression layer.

In addition, for the storage usage of CAFFE, we make constraints on reading input images from the given image dataset. In our deep CNN model allowing multiple scaled image sizes, the maximum number of region proposals is w (the width of the image) multiplied by h (the height of the image) multiplied by k (the number of anchors of a region proposal). This means that the region proposal network requires a considerable amount of memory space. Therefore, we limited the feeding of the number of input images in hidden layers to less than or equal to 20 simultaneous images. Other than CAFFE, we used a version of the Keras model, which uses Tensorflow and GPUs. This other implementation had similar results. Therefore, we will not discuss it in this paper.

For the training of our CNN model, we use model parameters given in [4], which are actually initialized by pre-training model parameters of ImageNet. We test 9000 images from datasets with PASCAL VOC 2007 and also tests with Microsoft’s COCO dataset. As we said before, we consider more only on the number of detected objects but not the improving or computational speeds. Therefore, we are pretty sure that our model training procedure is adequately fitted to our purpose.

For each region proposal from an input image, an object is randomly counted on, and we randomly sample 256 anchors in an image as a mini-batch size. Instead of considering the super-resolution method on selected region proposals instead of interpolation methods, any input images with variable sizes or shapes can be supported with better cropped or warped regions. But we do not consider this right now. We will leave this for future research. Instead, we processed with low-resolution images as the input dataset, processing super-resolution on these images, then detecting objects.

4. Performance Evaluations

To detect objects from images in CNNs, there are two kinds of popular file formats as input datasets. Those formats are JPG and Bitmap. However, we have considered several more image file formats as well as JPG and Bitmap formats to get better image quality for preprocessing and to get better detection objects during testing images. While datasets such as PASCAL VOC2007 or COCO are generated images with the JPG file format, in super-resolution processing, the JPG format did not seem to offer results as good as the Bitmap format. Such effects on preprocessing, we decided to use Bitmap format, and then we worked to convert the images from JPG format into BMP format before a preprocessing procedure. Also, we scaled down the image size, reducing each width and height by a factor of three to generate images of lower quality, instead of directly collecting cropped or warped images. Then, the image was further processed with bilinear and bicubic interpolations, as well as the super-resolution method. Thus, we built three different image datasets as preprocessing in our proposed method.

In our model, we implemented the preprocessing procedure based on [26]. Rather than training with nearly 50 images as in most of the published single image super-resolution models, we trained our model based on their initial parameters and with 100 PASCAL VOC2007 images and COCO images. We did not find any considerable overfitting with this number of training images. We generated images as the preprocessing of our method with randomly selected images from PASCAL VOC2007 [16][33], images from [34], and additionally with the COCO dataset images. Newly built images after our preprocessing are similar categories and the number of objects in an image. Therefore, we tested the object classification with 520 images randomly extracted from VOC2007 and 1224 images from COCOs MSO [35].

As shown in Figure 3, we knew that the dataset was annotated with a different number of objects compared to our intuition, and our proposed model found it. For example, the first image of Figure 3 was annotated with no number of any designated object, but our proposed model caught an object or several objects, as shown in the image. The second image of Figure 3 shows that a sofa was detected even though the image was annotated with no object. Because of these differences, we decided to use roughly the given annotations from the given dataset. A larger number of people than the labeled number was detected in the third image of Figure 3. This meant that we needed to manually count the number of objects in each image in the dataset to get the precision-recall measures.

4.1.  Comparison of Output Pictures

Even though some of the images from the PASCAL VOC 2007 were  detected as   the   image   had no   objects in the   designated

Figure 3: Some Images labeled with no objects on annotated dataset but detected objects by our method.

categories, multiple objects in the categories are detected from almost all of the images, as shown in the results given in Table 10. There were three different preprocessing models, namely bilinear interpolation, bicubic interpolation, and our super-resolution method. After the preprocessing of input images, our model for object detection was set with a learning rate of 0.001 to train and detection scores with 0.8 or higher compared with ground truths (IoU) at the testing stage. Our model has 21 categories that are randomly selected and which are independent of each other.

As shown in Figure 4, Figure 5, and Figure 6, green color bars (the first column) show the number of detected objects in each category as given by bilinear interpolation, blue color bars (the second column) show the number as given by bicubic interpolation, and red color bars (the third column) show the results for our proposed approach. Both the number of objects detected as y-coordinate and the detection rate as x-coordinate are shown.

As shown in Table 1, our model detects more objects than the other two models, but the average detection scores are not very different. This demonstrates that our model generates improved input images of which objects had lower scores than 0.8 compared with ground truth images in the other two models, which are then fed into the CNN which is able to increase the number of detections of categories. Here we mean the detection score is the probability that the detected object belongs to the class. Therefore, objects with near- or over-threshold scores (80% or above) are included even though they did not get the scores over the threshold value in the bilinear or bicubic interpolation models. Figure 4 shows the number of detected objects from the PASCAL VOC 2007 dataset and the comparison between them.

In Table 1, Table 2, In Figure 7, our proposed model detects a chair which is quite a small object compared with the image size, while the other two models did not detect this ‘chair’ object. It is looked like that a small change by preprocessing before CNN can make a big difference. Even though the other chair, which is bigger than the chair, is detected in all three models, our proposed model has a little higher score. Thus, we can conclude our model can help to improve image qualities in image analyses. In particular, with video surveillance camera systems, our model may help to detect more objects. Therefore, we are interested in surveillance camera systems as a future research topic.

#tp means the number of correctly predicted objects (true positive), and the mean column shows the classification average probabilities predicted.

Figure 4: Histogram for object detection models: images from VOC 2007 dataset.

With the dataset of Microsoft MSO as another image dataset for testing object classification, we present results in Figure 5 and Table 2. As with the VOC2007 dataset, our proposed model detects more objects than the two other models. This indicates that the image quality of low scored objects improved, leading to better detection. In other words, in our model, the image quality of low-scored objects improved, leading to detection scores (which is the probability that the object is the class) above 80%.

The main change in the COCO 2017 test dataset [36] is that instead of an 83K/41K train/validation split, the split is now 118K/5K for train/validation. Also, for testing, in 2017, the test set only had two splits (dev / challenge), instead of the four splits (dev / standard / reserve / challenge) used in previous years. For training of this dataset, we followed similar procedures as we did in PASCAL VOC 2007. For this dataset, we present the results in Figure 6 and In Figure 7, our proposed model detects a chair which is quite a small object compared with the image size, while the other two models did not detect this ‘chair’ object. It is looked like that a small change by preprocessing before CNN can make a big difference. Even though the other chair, which is bigger than the chair, is detected in all three models, our proposed model has a little higher score. Thus, we can conclude our model can help to improve image qualities in image analyses. In particular, with video surveillance camera systems, our model may help to detect more objects. Therefore, we are interested in surveillance camera systems as a future research topic.

. The total number of detected objects using our model was much bigger than the bilinear or bicubic models. However, the mean probabilities of detected objects were similar or a little bit lower. This indicates that many objects which were not detected with bilinear or bicubic interpolation methods were detected with our method.

Table 1: The number of detected objects on each category and the prediction scores on three different preprocessing and CNN with the PASCAL VOC2007 dataset.

Figure 5: Histogram for object detection models: images from the MSO dataset.

4.2.  Big vs. Small ROI Pictures and Their Detection Rates

As mentioned above, if an object is sufficiently large relative to the size of the image which contains the object in a region proposal, objects from interpolated images with bilinear or bicubic methods are identified satisfactorily and sometimes may have better perfor mance means that there is a higher probability that the object is identified as being in the specified class. However, the model we have proposed obtains much better results when the objects are from small bounding boxes, or if there is a small ratio of object size to the size of the image which contains the object (it means small anchor boxes in region proposals).

Table 2: The number of detected objects on each category and the prediction scores on three different preprocessing and CNN with the MSO dataset.

Figure 6: Histogram for object detection models: images from COCO 2017 dataset.

Figure 7: Examples of small object scales through three models

In Figure 7, our proposed model detects a chair which is quite a small object compared with the image size, while the other two models did not detect this ‘chair’ object. It is looked like that a small change by preprocessing before CNN can make a big difference. Even though the other chair, which is bigger than the chair, is detected in all three models, our proposed model has a little higher score. Thus, we can conclude our model can help to improve image qualities in image analyses. In particular, with video surveillance camera systems, our model may help to detect more objects. Therefore, we are interested in surveillance camera systems as a future research topic.

5. Conclusion

Our proposed model appears particularly powerful in three scenarios; firstly, where there are relatively small objects in large pictures; secondly, where there is warping in the region proposals approach; and finally, with object detection from cropped images. Commonly, there are many noisy video images from surveillance camera systems, especially with night vision systems. Our method will remove a significant amount of aliased or mosaicked areas in these images, and so help to detect more objects. We compared our scheme with other approaches on three datasets showing an increased object in each case.

In future work, we will implement the super-resolution method confined to the bounding box areas. This will reduce the needed computational resources and allow us to use this method in real-time processing [37] and achieve better object recognition in this application. We will demonstrate this capability using modern streaming software environments [38].

Table 3: The number of detected objects on each category and the prediction scores on three different preprocessing and CNN with the COCO 2017 dataset.

Acknowledgment

This work was supported by the 2017 sabbatical year research grant of the Korea Polytechnic University.

Geoffrey Fox was partially supported by NSF CIF21 DIBBS 1443054 and NSF CINES 1835598. We thank the Futuresystems group for the computational resources.

1. Bokyoon Na, Geoffrey Fox, “Object Detection by a Super-Resolution Method and a Convolutional Neural Networks”, IEEE International Conference on Big Data (Big Data) (2018)
2. Alex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton. “ImageNet Classification with Deep Convolution Neural Networks”. In Advances in Neural Information Processing Systems (NIPS), (1097-1105). (2012).
3. Girshick Ross. “Fast R-CNN”. arXiv: 1504.08083v2 [cs.CV] 27 Sep 2015.
4. Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. “Faster R-CNN: Towards Real-time Object Detection with Region Proposal Networks”. arXiv:1506.01497v3[cs.CV] 6 Jan 2016.
5. David Berthelot Schumm, Luke Metz Thomas. “BEGAN: Boundary Equilibrium Generative Adversarial Networks”. arXiv: 1703.10717v2 [cs.LG]. (2017).
6. Keiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. “Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recogniton”. arXiv:1406.4729v4. (2015).
7. Ross Girshick, Jeff Donahua, Trevor Darrell, Jitendra Malik, “Region-Based Convolution Networks for Accurate Object Detection and Segmentation”, IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 38, NO.1, 2016.
8. Karen Simonyan, Zisserman Andrew. “Very Deep Convolutional Networks for Large-Scale Image Recognition”. ICLR. (2015).
9. Joseph Redmon, Santosh Divvala, Ross Girshick, Ali Farhadi, “You Only Look Once: Unified, Real-time Object Detection”, The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779-788
10. Kaiming He, Georgia Gkioxari, Piotr Dollar, Ross Girshick, “Mask R-CNN”, The IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2961-2969
11. Ghosh Nachum and Debiprasad Ofir. https://www.quora.com. “What-are-Generative-Adversarial-Networks-GANs”.
12. Tim Salimans Goodfellow, Wojciech Zaremba, Vicki Cheung Ian. “Improved Techniques for Training GANs”. arXiv:1606.03498v1. (2016).
13. Takeru Miyato M Dai, Ian Goodfellow Andrew. “Adversarial Training Methods for Semi-Supervised Text Classification”. ICLR. (2017).
14. Ian J. Goodfellow Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courvill, Yoshua BengioJean. “Generative Adversarial Nets”. arXiv:1406.2661v1. (2014).
15. Sara Sabour, Nicholas Frosst, Geoffrey E. Hinton, “Dynamic Routing Between Capsules”, Advances in Neural Information Processing Systems 30 (NIPS 2017)
16. Everingham Van Gool, C. K. I. Williams, J. Winn, and A. ZissermanL. “The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results”. (2007).
17. Achanta R., Estrada F., Wils P., Süsstrunk S. “Salient Region Detection and Segmentation”. In: Gasteratos A., Vincze M., Tsotsos J.K. (eds) Computer Vision Systems. ICVS 2008. Lecture Notes in Computer Science, vol 5008. Springer, Berlin, Heidelberg. (2008)
18. Volodymyr Mnih, Nicolas Heess, Alex Graves, and Koray Kavukcuoglu. “Recurrent Models of Visual Attention”. arXiv:1406.6247v1[cs.LG] 24 Jun 2014.
19. Jimmy Lei Ba, Volodymyr Mnih, and Koray Kavukcuoglu. “Multiple Object Recognition with Visual Attention”. arXiv:1412.7755v2[cs.LG] 23 Apr 2015.
20. Karol Gregor Danihelka, Alex Graves, Danilo Jimenez Rezende, Daan Wierstra Ivo. “DRAW: A Recurrent Neural Network For Image Generation”. arXiv:1502.04623v2[cs.CV] 20 May 2015.
21. R.R. Uijlingsvan de Sande, T. Gevers, and A.W.M. Smeulders K.E.A. “Selective Search for Object Recognition”. IJCV. (2012).
22. Michael Elad, Arie Feuer, “Restoration of a Single Superresolution Image from Several Blurred, Noisy, and Undersampled Measured Images”, IEEE Transactions on Image Processing, Vol. 6, No. 12, December 1997
23. Jianchao Yang, John Wright, Thomas S. Huang, and Yi Ma, “Image Super-Resolution Via Sparse Representation”, IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 19, NO. 11, pp2861, NOVEMBER 2010
24. Nasrollahi Kamal, Guerrero Escalera Sergio, Rasti Pejman, Anbarjafari Gholamerza, Baro Xavier, J. Escalante Hugo, Moeslund B. Thomas. “Deep Learning based Super-Resolution for Improved Action Recognition”. In International Conference on Image Processing Theory, Tools and Applications (IPTA) IEEE Signal Processing Society. (2015).
25. Brian C. Tom, Aggelos K. Katsaggelos, Nikolas P. Galatsanos. “Reconstruction of a high resolution image from registration and restoration of low resolution images”. In Proceedings of IEEE International Conference on Image Processing, pages 553-557, 1994
26. Kai Zhang Wang, Wangmeng Zuo, Hongzhi Zhang, Lei ZhangBaoquan. “Joint Learning of Multiple Regressors for Single Image Super Resolution”. IEEE Singnal Processing Letters. Vol. 23 No. 1. (2016).
27. Zhaowen Wang Liu, Jianchao Yang, Wei Han, Thomas Huang Ding. “Deep Networks for Image Super Resolution with Sparse Prior”. ICCV. (2015).
28. Chao Dong Change Loy, Kaiming He, Xiaoou Tang Chen. “Image Super Resolution Using Deep Convolution Networks”. arXiv:1501.00092v3.v (2015).
29. Christian Ledig Theis, Ferenc Huszar, Jose Caballero, Andrew Cunningham, Alejandro Acosta, Andrew Aitken, Alykhan Tejani, Johannes Totz, Zehan Wang, Wenzhe ShiLucas. “Photo-Realistic Single Image Super Resolution Using a Generative Adversarial Network”. arXiv:1609.04802v5[cs.CV]. (2017).
30. Georgia Gkioxari, Ross Girshick, Piotr Dollár, Kaiming He, “Detecting and Recognizing Human-Object Interactions”, The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 8359-8367
31. Bishop Christopher. “Pattern Recognition and Machine Learning”. Springer. (2006).
32. Zhang Zhang, Baoquan Wang, Wangmeng Zuo, Hongzhi ZhangKai. “Joint Learning of Multiple Regressors for Single Image Super-Resolution”. IEEE Signal processing letters, Vol. 23, No.1. (2016).
33. Olga Russakovsky Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg and Li Fei-Fei Jia. “ImageNet Large Scale Visual Recognition Challenge”. arXiv:1409.0575. (2014).
34. Olga Russakovsky Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg and Li Fei-Fei Jia, “ImageNet Large Scale Visual Recognition Challenge 2017″. (2017).
35. Zhang Ma, Shuga Sameki, Mehrnoosh Sclaroff, Stan Betke, Margrit Lin, Zhe Shen, Xiaohui Price, Brian Much, Radom Jianming. “Salient Object Subitizing”. IEEE Conference on Computer Vision and Pattern Recognition (CVPR). (2015).
36. Tsung-Yi Lin, Michael Maire, Serge Belongie, Lubomir Bourdev, Ross Girshick, James Hays, Pietro Perona, Deva Ramanan, C. Lawrence Zitnick, Piotr Dollár. “Microsoft COCO: Common Objects in Context”, arXiv:1405.0312 [cs.CV]
37. Xinyuan Huang, Geoffrey C. Fox, Sergey Serebryakov, Ankur Mohan, Pawel Morkisz, Debojyoti Dutta, “Benchmarking Deep Learning for Time Series: Challenges and Directions“, Stream-ML workshopat IEEE Big Data Conference, Los Angeles CA December 10, 2019
38. Vibhatha Abeykoon, Supun Kamburugamuve, Kannan Govindrarajan,  Pulasthi Wickramasinghe, Chathura Widanage, Niranda Perera, Ahmet Uyar, Gurhan Gunduz, Selahattin Akkas, Gregor Von Laszewski, and Geoffrey Fox, “Streaming Machine Learning Algorithms with Big Data Systems”, Stream-ML workshopat IEEE Big Data Conference, Los Angeles CA December 10, 2019

Citations by Dimensions

Citations by PlumX

Google Scholar

Crossref Citations

1. Vikas Thammanna Gowda, Landis Humphrey, Aiden Kadoch, YinBo Chen, Olivia Roberts, "Multi Attribute Stratified Sampling: An Automated Framework for Privacy-Preserving Healthcare Data Publishing with Multiple Sensitive Attributes", Advances in Science, Technology and Engineering Systems Journal, vol. 11, no. 1, pp. 51–68, 2026. doi: 10.25046/aj110106
2. Kohinur Parvin, Eshat Ahmad Shuvo, Wali Ashraf Khan, Sakibul Alam Adib, Tahmina Akter Eiti, Mohammad Shovon, Shoeb Akter Nafiz, "Computationally Efficient Explainable AI Framework for Skin Cancer Detection", Advances in Science, Technology and Engineering Systems Journal, vol. 11, no. 1, pp. 11–24, 2026. doi: 10.25046/aj110102
3. David Degbor, Haiping Xu, Pratiksha Singh, Shannon Gibbs, Donghui Yan, "StradNet: Automated Structural Adaptation for Efficient Deep Neural Network Design", Advances in Science, Technology and Engineering Systems Journal, vol. 10, no. 6, pp. 29–41, 2025. doi: 10.25046/aj100603
4. Jenna Snead, Nisa Soltani, Mia Wang, Joe Carson, Bailey Williamson, Kevin Gainey, Stanley McAfee, Qian Zhang, "3D Facial Feature Tracking with Multimodal Depth Fusion", Advances in Science, Technology and Engineering Systems Journal, vol. 10, no. 5, pp. 11–19, 2025. doi: 10.25046/aj100502
5. Glender Brás, Samara Leal, Breno Sousa, Gabriel Paes, Cleberson Junior, João Souza, Rafael Assis, Tamires Marques, Thiago Teles Calazans Silva, "Machine Learning Methods for University Student Performance Prediction in Basic Skills based on Psychometric Profile", Advances in Science, Technology and Engineering Systems Journal, vol. 10, no. 4, pp. 1–13, 2025. doi: 10.25046/aj100401
6. khawla Alhasan, "Predictive Analytics in Marketing: Evaluating its Effectiveness in Driving Customer Engagement", Advances in Science, Technology and Engineering Systems Journal, vol. 10, no. 3, pp. 45–51, 2025. doi: 10.25046/aj100306
7. Surapol Vorapatratorn, Nontawat Thongsibsong, "AI-Based Photography Assessment System using Convolutional Neural Networks", Advances in Science, Technology and Engineering Systems Journal, vol. 10, no. 2, pp. 28–34, 2025. doi: 10.25046/aj100203
8. Khalifa Sylla, Birahim Babou, Mama Amar, Samuel Ouya, "Impact of Integrating Chatbots into Digital Universities Platforms on the Interactions between the Learner and the Educational Content", Advances in Science, Technology and Engineering Systems Journal, vol. 10, no. 1, pp. 13–19, 2025. doi: 10.25046/aj100103
9. Ahmet Emin Ünal, Halit Boyar, Burcu Kuleli Pak, Vehbi Çağrı Güngör, "Utilizing 3D models for the Prediction of Work Man-Hour in Complex Industrial Products using Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 6, pp. 01–11, 2024. doi: 10.25046/aj090601
10. Haruki Murakami, Takuma Miwa, Kosuke Shima, Takanobu Otsuka, "Proposal and Implementation of Seawater Temperature Prediction Model using Transfer Learning Considering Water Depth Differences", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 4, pp. 01–06, 2024. doi: 10.25046/aj090401
11. Brandon Wetzel, Haiping Xu, "Deploying Trusted and Immutable Predictive Models on a Public Blockchain Network", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 3, pp. 72–83, 2024. doi: 10.25046/aj090307
12. Anirudh Mazumder, Kapil Panda, "Leveraging Machine Learning for a Comprehensive Assessment of PFAS Nephrotoxicity", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 3, pp. 62–71, 2024. doi: 10.25046/aj090306
13. Taichi Ito, Ken’ichi Minamino, Shintaro Umeki, "Visualization of the Effect of Additional Fertilization on Paddy Rice by Time-Series Analysis of Vegetation Indices using UAV and Minimizing the Number of Monitoring Days for its Workload Reduction", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 3, pp. 29–40, 2024. doi: 10.25046/aj090303
14. Henry Toal, Michelle Wilber, Getu Hailu, Arghya Kusum Das, "Evaluation of Various Deep Learning Models for Short-Term Solar Forecasting in the Arctic using a Distributed Sensor Network", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 3, pp. 12–28, 2024. doi: 10.25046/aj090302
15. Tinofirei Museba, Koenraad Vanhoof, "An Adaptive Heterogeneous Ensemble Learning Model for Credit Card Fraud Detection", Advances in Science, Technology and Engineering Systems Journal, vol. 9, no. 3, pp. 01–11, 2024. doi: 10.25046/aj090301
16. Toya Acharya, Annamalai Annamalai, Mohamed F Chouikha, "Optimizing the Performance of Network Anomaly Detection Using Bidirectional Long Short-Term Memory (Bi-LSTM) and Over-sampling for Imbalance Network Traffic Data", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 6, pp. 144–154, 2023. doi: 10.25046/aj080614
17. Renhe Chi, "Comparative Study of J48 Decision Tree and CART Algorithm for Liver Cancer Symptom Analysis Using Data from Carnegie Mellon University", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 6, pp. 57–64, 2023. doi: 10.25046/aj080607
18. Ng Kah Kit, Hafeez Ullah Amin, Kher Hui Ng, Jessica Price, Ahmad Rauf Subhani, "EEG Feature Extraction based on Fast Fourier Transform and Wavelet Analysis for Classification of Mental Stress Levels using Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 6, pp. 46–56, 2023. doi: 10.25046/aj080606
19. Kitipoth Wasayangkool, Kanabadee Srisomboon, Chatree Mahatthanajatuphat, Wilaiporn Lee, "Accuracy Improvement-Based Wireless Sensor Estimation Technique with Machine Learning Algorithms for Volume Estimation on the Sealed Box", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 3, pp. 108–117, 2023. doi: 10.25046/aj080313
20. Chaiyaporn Khemapatapan, Thammanoon Thepsena, "Forecasting the Weather behind Pa Sak Jolasid Dam using Quantum Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 3, pp. 54–62, 2023. doi: 10.25046/aj080307
21. Der-Jiun Pang, "Hybrid Machine Learning Model Performance in IT Project Cost and Duration Prediction", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 2, pp. 108–115, 2023. doi: 10.25046/aj080212
22. Paulo Gustavo Quinan, Issa Traoré, Isaac Woungang, Ujwal Reddy Gondhi, Chenyang Nie, "Hybrid Intrusion Detection Using the AEN Graph Model", Advances in Science, Technology and Engineering Systems Journal, vol. 8, no. 2, pp. 44–63, 2023. doi: 10.25046/aj080206
23. Ossama Embarak, "Multi-Layered Machine Learning Model For Mining Learners Academic Performance", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 850–861, 2021. doi: 10.25046/aj060194
24. Anh-Thu Mai, Duc-Huy Nguyen, Thanh-Tin Dang, "Transfer and Ensemble Learning in Real-time Accurate Age and Age-group Estimation", Advances in Science, Technology and Engineering Systems Journal, vol. 7, no. 6, pp. 262–268, 2022. doi: 10.25046/aj070630
25. Roy D Gregori Ayon, Md. Sanaullah Rabbi, Umme Habiba, Maoyejatun Hasana, "Bangla Speech Emotion Detection using Machine Learning Ensemble Methods", Advances in Science, Technology and Engineering Systems Journal, vol. 7, no. 6, pp. 70–76, 2022. doi: 10.25046/aj070608
26. Deeptaanshu Kumar, Ajmal Thanikkal, Prithvi Krishnamurthy, Xinlei Chen, Pei Zhang, "Analysis of Different Supervised Machine Learning Methods for Accelerometer-Based Alcohol Consumption Detection from Physical Activity", Advances in Science, Technology and Engineering Systems Journal, vol. 7, no. 4, pp. 147–154, 2022. doi: 10.25046/aj070419
27. Zhumakhan Nazir, Temirlan Zarymkanov, Jurn-Guy Park, "A Machine Learning Model Selection Considering Tradeoffs between Accuracy and Interpretability", Advances in Science, Technology and Engineering Systems Journal, vol. 7, no. 4, pp. 72–78, 2022. doi: 10.25046/aj070410
28. Ayoub Benchabana, Mohamed-Khireddine Kholladi, Ramla Bensaci, Belal Khaldi, "A Supervised Building Detection Based on Shadow using Segmentation and Texture in High-Resolution Images", Advances in Science, Technology and Engineering Systems Journal, vol. 7, no. 3, pp. 166–173, 2022. doi: 10.25046/aj070319
29. Osaretin Eboya, Julia Binti Juremi, "iDRP Framework: An Intelligent Malware Exploration Framework for Big Data and Internet of Things (IoT) Ecosystem", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 5, pp. 185–202, 2021. doi: 10.25046/aj060521
30. Arwa Alghamdi, Graham Healy, Hoda Abdelhafez, "Machine Learning Algorithms for Real Time Blind Audio Source Separation with Natural Language Detection", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 5, pp. 125–140, 2021. doi: 10.25046/aj060515
31. Baida Ouafae, Louzar Oumaima, Ramdi Mariam, Lyhyaoui Abdelouahid, "Survey on Novelty Detection using Machine Learning Techniques", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 5, pp. 73–82, 2021. doi: 10.25046/aj060510
32. Radwan Qasrawi, Stephanny VicunaPolo, Diala Abu Al-Halawa, Sameh Hallaq, Ziad Abdeen, "Predicting School Children Academic Performance Using Machine Learning Techniques", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 5, pp. 08–15, 2021. doi: 10.25046/aj060502
33. Zhiyuan Chen, Howe Seng Goh, Kai Ling Sin, Kelly Lim, Nicole Ka Hei Chung, Xin Yu Liew, "Automated Agriculture Commodity Price Prediction System with Machine Learning Techniques", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 4, pp. 376–384, 2021. doi: 10.25046/aj060442
34. Hathairat Ketmaneechairat, Maleerat Maliyaem, Chalermpong Intarat, "Kamphaeng Saen Beef Cattle Identification Approach using Muzzle Print Image", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 4, pp. 110–122, 2021. doi: 10.25046/aj060413
35. Md Mahmudul Hasan, Nafiul Hasan, Dil Afroz, Ferdaus Anam Jibon, Md. Arman Hossen, Md. Shahrier Parvage, Jakaria Sulaiman Aongkon, "Electroencephalogram Based Medical Biometrics using Machine Learning: Assessment of Different Color Stimuli", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 3, pp. 27–34, 2021. doi: 10.25046/aj060304
36. Dominik Štursa, Daniel Honc, Petr Doležel, "Efficient 2D Detection and Positioning of Complex Objects for Robotic Manipulation Using Fully Convolutional Neural Network", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 915–920, 2021. doi: 10.25046/aj0602104
37. Kenza Aitelkadi, Hicham Outmghoust, Salahddine laarab, Kaltoum Moumayiz, Imane Sebari, "Detection and Counting of Fruit Trees from RGB UAV Images by Convolutional Neural Networks Approach", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 887–893, 2021. doi: 10.25046/aj0602101
38. Md Mahmudul Hasan, Nafiul Hasan, Mohammed Saud A Alsubaie, "Development of an EEG Controlled Wheelchair Using Color Stimuli: A Machine Learning Based Approach", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 754–762, 2021. doi: 10.25046/aj060287
39. Antoni Wibowo, Inten Yasmina, Antoni Wibowo, "Food Price Prediction Using Time Series Linear Ridge Regression with The Best Damping Factor", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 694–698, 2021. doi: 10.25046/aj060280
40. Javier E. Sánchez-Galán, Fatima Rangel Barranco, Jorge Serrano Reyes, Evelyn I. Quirós-McIntire, José Ulises Jiménez, José R. Fábrega, "Using Supervised Classification Methods for the Analysis of Multi-spectral Signatures of Rice Varieties in Panama", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 552–558, 2021. doi: 10.25046/aj060262
41. Phillip Blunt, Bertram Haskins, "A Model for the Application of Automatic Speech Recognition for Generating Lesson Summaries", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 526–540, 2021. doi: 10.25046/aj060260
42. Sebastianus Bara Primananda, Sani Muhamad Isa, "Forecasting Gold Price in Rupiah using Multivariate Analysis with LSTM and GRU Neural Networks", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 2, pp. 245–253, 2021. doi: 10.25046/aj060227
43. Byeongwoo Kim, Jongkyu Lee, "Fault Diagnosis and Noise Robustness Comparison of Rotating Machinery using CWT and CNN", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 1279–1285, 2021. doi: 10.25046/aj0601146
44. Alisson Steffens Henrique, Anita Maria da Rocha Fernandes, Rodrigo Lyra, Valderi Reis Quietinho Leithardt, Sérgio D. Correia, Paul Crocker, Rudimar Luis Scaranto Dazzi, "Classifying Garments from Fashion-MNIST Dataset Through CNNs", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 989–994, 2021. doi: 10.25046/aj0601109
45. Md Mahmudul Hasan, Nafiul Hasan, Mohammed Saud A Alsubaie, Md Mostafizur Rahman Komol, "Diagnosis of Tobacco Addiction using Medical Signal: An EEG-based Time-Frequency Domain Analysis Using Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 842–849, 2021. doi: 10.25046/aj060193
46. Reem Bayari, Ameur Bensefia, "Text Mining Techniques for Cyberbullying Detection: State of the Art", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 783–790, 2021. doi: 10.25046/aj060187
47. Inna Valieva, Iurii Voitenko, Mats Björkman, Johan Åkerberg, Mikael Ekström, "Multiple Machine Learning Algorithms Comparison for Modulation Type Classification Based on Instantaneous Values of the Time Domain Signal and Time Series Statistics Derived from Wavelet Transform", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 658–671, 2021. doi: 10.25046/aj060172
48. Carlos López-Bermeo, Mauricio González-Palacio, Lina Sepúlveda-Cano, Rubén Montoya-Ramírez, César Hidalgo-Montoya, "Comparison of Machine Learning Parametric and Non-Parametric Techniques for Determining Soil Moisture: Case Study at Las Palmas Andean Basin", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 636–650, 2021. doi: 10.25046/aj060170
49. Ndiatenda Ndou, Ritesh Ajoodha, Ashwini Jadhav, "A Case Study to Enhance Student Support Initiatives Through Forecasting Student Success in Higher-Education", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 230–241, 2021. doi: 10.25046/aj060126
50. Lonia Masangu, Ashwini Jadhav, Ritesh Ajoodha, "Predicting Student Academic Performance Using Data Mining Techniques", Advances in Science, Technology and Engineering Systems Journal, vol. 6, no. 1, pp. 153–163, 2021. doi: 10.25046/aj060117
51. Sara Ftaimi, Tomader Mazri, "Handling Priority Data in Smart Transportation System by using Support Vector Machine Algorithm", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 6, pp. 1422–1427, 2020. doi: 10.25046/aj0506172
52. Othmane Rahmaoui, Kamal Souali, Mohammed Ouzzif, "Towards a Documents Processing Tool using Traceability Information Retrieval and Content Recognition Through Machine Learning in a Big Data Context", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 6, pp. 1267–1277, 2020. doi: 10.25046/aj0506151
53. Puttakul Sakul-Ung, Amornvit Vatcharaphrueksadee, Pitiporn Ruchanawet, Kanin Kearpimy, Hathairat Ketmaneechairat, Maleerat Maliyaem, "Overmind: A Collaborative Decentralized Machine Learning Framework", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 6, pp. 280–289, 2020. doi: 10.25046/aj050634
54. Andrea Generosi, Silvia Ceccacci, Samuele Faggiano, Luca Giraldi, Maura Mengoni, "A Toolkit for the Automatic Analysis of Human Behavior in HCI Applications in the Wild", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 6, pp. 185–192, 2020. doi: 10.25046/aj050622
55. Pamela Zontone, Antonio Affanni, Riccardo Bernardini, Leonida Del Linz, Alessandro Piras, Roberto Rinaldo, "Supervised Learning Techniques for Stress Detection in Car Drivers", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 6, pp. 22–29, 2020. doi: 10.25046/aj050603
56. Sherif H. ElGohary, Aya Lithy, Shefaa Khamis, Aya Ali, Aya Alaa el-din, Hager Abd El-Azim, "Interactive Virtual Rehabilitation for Aphasic Arabic-Speaking Patients", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 1225–1232, 2020. doi: 10.25046/aj0505148
57. Kodai Kitagawa, Koji Matsumoto, Kensuke Iwanaga, Siti Anom Ahmad, Takayuki Nagasaki, Sota Nakano, Mitsumasa Hida, Shogo Okamatsu, Chikamune Wada, "Posture Recognition Method for Caregivers during Postural Change of a Patient on a Bed using Wearable Sensors", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 1093–1098, 2020. doi: 10.25046/aj0505133
58. Daniyar Nurseitov, Kairat Bostanbekov, Maksat Kanatov, Anel Alimova, Abdelrahman Abdallah, Galymzhan Abdimanap, "Classification of Handwritten Names of Cities and Handwritten Text Recognition using Various Deep Learning Models", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 934–943, 2020. doi: 10.25046/aj0505114
59. Khalid A. AlAfandy, Hicham Omara, Mohamed Lazaar, Mohammed Al Achhab, "Using Classic Networks for Classifying Remote Sensing Images: Comparative Study", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 770–780, 2020. doi: 10.25046/aj050594
60. Khalid A. AlAfandy, Hicham, Mohamed Lazaar, Mohammed Al Achhab, "Investment of Classic Deep CNNs and SVM for Classifying Remote Sensing Images", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 652–659, 2020. doi: 10.25046/aj050580
61. Rajesh Kumar, Geetha S, "Malware Classification Using XGboost-Gradient Boosted Decision Tree", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 536–549, 2020. doi: 10.25046/aj050566
62. Nghia Duong-Trung, Nga Quynh Thi Tang, Xuan Son Ha, "Interpretation of Machine Learning Models for Medical Diagnosis", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 469–477, 2020. doi: 10.25046/aj050558
63. Oumaima Terrada, Soufiane Hamida, Bouchaib Cherradi, Abdelhadi Raihani, Omar Bouattane, "Supervised Machine Learning Based Medical Diagnosis Support System for Prediction of Patients with Heart Disease", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 269–277, 2020. doi: 10.25046/aj050533
64. Haytham Azmi, "FPGA Acceleration of Tree-based Learning Algorithms", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 237–244, 2020. doi: 10.25046/aj050529
65. Hicham Moujahid, Bouchaib Cherradi, Oussama El Gannour, Lhoussain Bahatti, Oumaima Terrada, Soufiane Hamida, "Convolutional Neural Network Based Classification of Patients with Pneumonia using X-ray Lung Images", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 5, pp. 167–175, 2020. doi: 10.25046/aj050522
66. Young-Jin Park, Hui-Sup Cho, "A Method for Detecting Human Presence and Movement Using Impulse Radar", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 770–775, 2020. doi: 10.25046/aj050491
67. Nghia Duong-Trung, Luyl-Da Quach, Chi-Ngon Nguyen, "Towards Classification of Shrimp Diseases Using Transferred Convolutional Neural Networks", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 724–732, 2020. doi: 10.25046/aj050486
68. Anouar Bachar, Noureddine El Makhfi, Omar EL Bannay, "Machine Learning for Network Intrusion Detection Based on SVM Binary Classification Model", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 638–644, 2020. doi: 10.25046/aj050476
69. Adonis Santos, Patricia Angela Abu, Carlos Oppus, Rosula Reyes, "Real-Time Traffic Sign Detection and Recognition System for Assistive Driving", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 600–611, 2020. doi: 10.25046/aj050471
70. Amar Choudhary, Deependra Pandey, Saurabh Bhardwaj, "Overview of Solar Radiation Estimation Techniques with Development of Solar Radiation Model Using Artificial Neural Network", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 589–593, 2020. doi: 10.25046/aj050469
71. Maroua Abdellaoui, Dounia Daghouj, Mohammed Fattah, Younes Balboul, Said Mazer, Moulhime El Bekkali, "Artificial Intelligence Approach for Target Classification: A State of the Art", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 445–456, 2020. doi: 10.25046/aj050453
72. Roberta Avanzato, Francesco Beritelli, "A CNN-based Differential Image Processing Approach for Rainfall Classification", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 438–444, 2020. doi: 10.25046/aj050452
73. Shahab Pasha, Jan Lundgren, Christian Ritz, Yuexian Zou, "Distributed Microphone Arrays, Emerging Speech and Audio Signal Processing Platforms: A Review", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 331–343, 2020. doi: 10.25046/aj050439
74. Ilias Kalathas, Michail Papoutsidakis, Chistos Drosos, "Optimization of the Procedures for Checking the Functionality of the Greek Railways: Data Mining and Machine Learning Approach to Predict Passenger Train Immobilization", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 287–295, 2020. doi: 10.25046/aj050435
75. Yosaphat Catur Widiyono, Sani Muhamad Isa, "Utilization of Data Mining to Predict Non-Performing Loan", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 4, pp. 252–256, 2020. doi: 10.25046/aj050431
76. Hai Thanh Nguyen, Nhi Yen Kim Phan, Huong Hoang Luong, Trung Phuoc Le, Nghi Cong Tran, "Efficient Discretization Approaches for Machine Learning Techniques to Improve Disease Classification on Gut Microbiome Composition Data", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 3, pp. 547–556, 2020. doi: 10.25046/aj050368
77. Ruba Obiedat, "Risk Management: The Case of Intrusion Detection using Data Mining Techniques", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 3, pp. 529–535, 2020. doi: 10.25046/aj050365
78. Krina B. Gabani, Mayuri A. Mehta, Stephanie Noronha, "Racial Categorization Methods: A Survey", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 3, pp. 388–401, 2020. doi: 10.25046/aj050350
79. Dennis Luqman, Sani Muhamad Isa, "Machine Learning Model to Identify the Optimum Database Query Execution Platform on GPU Assisted Database", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 3, pp. 214–225, 2020. doi: 10.25046/aj050328
80. Gillala Rekha, Shaveta Malik, Amit Kumar Tyagi, Meghna Manoj Nair, "Intrusion Detection in Cyber Security: Role of Machine Learning and Data Mining in Cyber Security", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 3, pp. 72–81, 2020. doi: 10.25046/aj050310
81. Ahmed EL Orche, Mohamed Bahaj, "Approach to Combine an Ontology-Based on Payment System with Neural Network for Transaction Fraud Detection", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 2, pp. 551–560, 2020. doi: 10.25046/aj050269
82. Lenin G. Falconi, Maria Perez, Wilbert G. Aguilar, Aura Conci, "Transfer Learning and Fine Tuning in Breast Mammogram Abnormalities Classification on CBIS-DDSM Database", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 2, pp. 154–165, 2020. doi: 10.25046/aj050220
83. Johannes Linden, Xutao Wang, Stefan Forsstrom, Tingting Zhang, "Productify News Article Classification Model with Sagemaker", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 2, pp. 13–18, 2020. doi: 10.25046/aj050202
84. Michael Wenceslaus Putong, Suharjito, "Classification Model of Contact Center Customers Emails Using Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 5, no. 1, pp. 174–182, 2020. doi: 10.25046/aj050123
85. Rehan Ullah Khan, Ali Mustafa Qamar, Mohammed Hadwan, "Quranic Reciter Recognition: A Machine Learning Approach", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 6, pp. 173–176, 2019. doi: 10.25046/aj040621
86. Mehdi Guessous, Lahbib Zenkouar, "An ML-optimized dRRM Solution for IEEE 802.11 Enterprise Wlan Networks", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 6, pp. 19–31, 2019. doi: 10.25046/aj040603
87. Toshiyasu Kato, Yuki Terawaki, Yasushi Kodama, Teruhiko Unoki, Yasushi Kambayashi, "Estimating Academic results from Trainees’ Activities in Programming Exercises Using Four Types of Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 5, pp. 321–326, 2019. doi: 10.25046/aj040541
88. Nindhia Hutagaol, Suharjito, "Predictive Modelling of Student Dropout Using Ensemble Classifier Method in Higher Education", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 4, pp. 206–211, 2019. doi: 10.25046/aj040425
89. Fernando Hernández, Roberto Vega, Freddy Tapia, Derlin Morocho, Walter Fuertes, "Early Detection of Alzheimer’s Using Digital Image Processing Through Iridology, An Alternative Method", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 3, pp. 126–137, 2019. doi: 10.25046/aj040317
90. Abba Suganda Girsang, Andi Setiadi Manalu, Ko-Wei Huang, "Feature Selection for Musical Genre Classification Using a Genetic Algorithm", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 2, pp. 162–169, 2019. doi: 10.25046/aj040221
91. Ajees Arimbassery Pareed, Sumam Mary Idicula, "A Relation Extraction System for Indian Languages", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 2, pp. 65–69, 2019. doi: 10.25046/aj040208
92. Konstantin Mironov, Ruslan Gayanov, Dmiriy Kurennov, "Observing and Forecasting the Trajectory of the Thrown Body with use of Genetic Programming", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 1, pp. 248–257, 2019. doi: 10.25046/aj040124
93. Bok Gyu Han, Hyeon Seok Yang, Ho Gyeong Lee, Young Shik Moon, "Low Contrast Image Enhancement Using Convolutional Neural Network with Simple Reflection Model", Advances in Science, Technology and Engineering Systems Journal, vol. 4, no. 1, pp. 159–164, 2019. doi: 10.25046/aj040115
94. Zheng Xie, Chaitanya Gadepalli, Farideh Jalalinajafabadi, Barry M.G. Cheetham, Jarrod J. Homer, "Machine Learning Applied to GRBAS Voice Quality Assessment", Advances in Science, Technology and Engineering Systems Journal, vol. 3, no. 6, pp. 329–338, 2018. doi: 10.25046/aj030641
95. Richard Osei Agjei, Emmanuel Awuni Kolog, Daniel Dei, Juliet Yayra Tengey, "Emotional Impact of Suicide on Active Witnesses: Predicting with Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 3, no. 5, pp. 501–509, 2018. doi: 10.25046/aj030557
96. Sudipta Saha, Aninda Saha, Zubayr Khalid, Pritam Paul, Shuvam Biswas, "A Machine Learning Framework Using Distinctive Feature Extraction for Hand Gesture Recognition", Advances in Science, Technology and Engineering Systems Journal, vol. 3, no. 5, pp. 72–81, 2018. doi: 10.25046/aj030510
97. Charles Frank, Asmail Habach, Raed Seetan, Abdullah Wahbeh, "Predicting Smoking Status Using Machine Learning Algorithms and Statistical Analysis", Advances in Science, Technology and Engineering Systems Journal, vol. 3, no. 2, pp. 184–189, 2018. doi: 10.25046/aj030221
98. Sehla Loussaief, Afef Abdelkrim, "Machine Learning framework for image classification", Advances in Science, Technology and Engineering Systems Journal, vol. 3, no. 1, pp. 1–10, 2018. doi: 10.25046/aj030101
99. Ruijian Zhang, Deren Li, "Applying Machine Learning and High Performance Computing to Water Quality Assessment and Prediction", Advances in Science, Technology and Engineering Systems Journal, vol. 2, no. 6, pp. 285–289, 2017. doi: 10.25046/aj020635
100. Batoul Haidar, Maroun Chamoun, Ahmed Serhrouchni, "A Multilingual System for Cyberbullying Detection: Arabic Content Detection using Machine Learning", Advances in Science, Technology and Engineering Systems Journal, vol. 2, no. 6, pp. 275–284, 2017. doi: 10.25046/aj020634