Knowing student learning styles represents an effective way to design the most suitable methodology for our students so that performance can improve with less effort for both students and teachers. ...However, a methodology is usually set in teaching guides according to the previous academic year's information without any knowledge of our current audience. In this work, a new software for learning styles and grade analysis based on the Honey‐Alonso Learning Styles Questionnaire has been proposed. This tool proposes the average learning style profiles of a given course by clustering student learning styles and analyzes the possible relation between grades and learning style profiles. By using that program, three different courses from Computer Sciences Engineering degrees during an academic year have been analyzed. The obtained results in our specific context exhibit that possible relation. This information could be useful to understand how students approach learning materials.
Full text
Available for:
BFBNIB, FZAB, GIS, IJS, KILJ, NLZOH, NUK, OILJ, SBCE, SBMB, UL, UM, UPUK
Breast cancer is the second most common cancer worldwide, primarily affecting women, while histopathological image analysis is one of the possibile methods used to determine tumor malignancy. ...Regarding image analysis, the application of deep learning has become increasingly prevalent in recent years. However, a significant issue is the unbalanced nature of available datasets, with some classes having more images than others, which may impact the performance of the models due to poorer generalizability. A possible strategy to avoid this problem is downsampling the class with the most images to create a balanced dataset. Nevertheless, this approach is not recommended for small datasets as it can lead to poor model performance. Instead, techniques such as data augmentation are traditionally used to address this issue. These techniques apply simple transformations such as translation or rotation to the images to increase variability in the dataset. Another possibility is using generative adversarial networks (GANs), which can generate images from a relatively small training set. This work aims to enhance model performance in classifying histopathological images by applying data augmentation using GANs instead of traditional techniques.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
A key factor in the fight against viral diseases such as the coronavirus (COVID-19) is the identification of virus carriers as early and quickly as possible, in a cheap and efficient manner. The ...application of deep learning for image classification of chest X-ray images of COVID-19 patients could become a useful pre-diagnostic detection methodology. However, deep learning architectures require large labelled datasets. This is often a limitation when the subject of research is relatively new as in the case of the virus outbreak, where dealing with small labelled datasets is a challenge. Moreover, in such context, the datasets are also highly imbalanced, with few observations from positive cases of the new disease. In this work we evaluate the performance of the semi-supervised deep learning architecture known as MixMatch with a very limited number of labelled observations and highly imbalanced labelled datasets. We demonstrate the critical impact of data imbalance to the model’s accuracy. Therefore, we propose a simple approach for correcting data imbalance, by re-weighting each observation in the loss function, giving a higher weight to the observations corresponding to the under-represented class. For unlabelled observations, we use the pseudo and augmented labels calculated by MixMatch to choose the appropriate weight. The proposed method improved classification accuracy by up to 18%, with respect to the non balanced MixMatch algorithm. We tested our proposed approach with several available datasets using 10, 15 and 20 labelled observations, for binary classification (COVID-19 positive and normal cases). For multi-class classification (COVID-19 positive, pneumonia and normal cases), we tested 30, 50, 70 and 90 labelled observations. Additionally, a new dataset is included among the tested datasets, composed of chest X-ray images of Costa Rican adult patients.
•COVID-19 detection deep learning architectures typically need many labels.•Also the datasets at the beginning of a virus outbreak are highly imbalanced.•Semi supervised data can be used to increase model’s accuracy with few labels.•The effect of data imbalance on semi-supervised learning is under-explored.•A method to correct data imbalance for semi supervised learning is proposed.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
The application of deep learning to image and video processing has become increasingly popular nowadays. Employing well-known pre-trained neural networks for detecting and classifying objects in ...images is beneficial in a wide range of application fields. However, diverse impediments may degrade the performance achieved by those neural networks. Particularly, Gaussian noise and brightness, among others, may be presented on images as sensor noise due to the limitations of image acquisition devices. In this work, we study the effect of the most representative noise types and brightness alterations on images in the performance of several state-of-the-art object detectors, such as YOLO or Faster-RCNN. Different experiments have been carried out and the results demonstrate how these adversities deteriorate their performance. Moreover, it is found that the size of objects to be detected is a factor that, together with noise and brightness factors, has a considerable impact on their performance.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
In this work we implement a COVID-19 infection detection system based on chest X-ray images with uncertainty estimation. Uncertainty estimation is vital for safe usage of computer aided diagnosis ...tools in medical applications. Model estimations with high uncertainty should be carefully analyzed by a trained radiologist. We aim to improve uncertainty estimations using unlabelled data through the MixMatch semi-supervised framework. We test popular uncertainty estimation approaches, comprising Softmax scores, Monte-Carlo dropout and deterministic uncertainty quantification. To compare the reliability of the uncertainty estimates, we propose the usage of the Jensen-Shannon distance between the uncertainty distributions of correct and incorrect estimations. This metric is statistically relevant, unlike most previously used metrics, which often ignore the distribution of the uncertainty estimations. Our test results show a significant improvement in uncertainty estimates when using unlabelled data. The best results are obtained with the use of the Monte Carlo dropout method.
Ensemble learning has demonstrated its efficiency in many computer vision tasks. In this paper, we address this paradigm within content based image retrieval (CBIR). We propose to build an ensemble ...of convolutional neural networks (CNNs), either by training the CNNs on different bags of images, or by using CNNs trained on the same dataset, but having different architectures. Each network is used to extract the class probability vectors from images to use them as representations. The final image representation is then generated by combining the extracted class probability vectors from the built ensemble. We show that the use of CNN ensembles is very efficient in generating a powerful image representation compared to individual CNNs. Moreover, we propose an Averarge Query Expansion technique for our proposal to enhance the retrieval results. Several experiments were conducted to extensively evaluate the application of ensemble learning in CBIR. Results in terms of precision, recall, and mean average precision show the outperformance of our proposal compared to the state of the art.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
•Most of foreground detection methods are pixel-based which implies higher complexity.•Several effective methods hardly reach real time execution for 320 × 240 pixels videos.•The rise of low cost ...devices requires lighter foreground detection algorithms.•Each scene has been represented with a set of tilings.•The final foreground mask consists of a combination of the masks of this set.
In this work a novel region-based approach for the detection of foreground in video sequences is presented. The model consists of an ensemble of layers or tilings, where each tiling represents, by means of randomly chosen parallelogram regions, the background of the scene. Currently, the image size of video surveillance cameras far exceeds one megapixel (more than 1024 × 768), and pixel-based proposals are poorly suited for near real-time ratios. Therefore, the analysis by pixel is replaced by an analysis by region, improving the final resolution by overlapping regions or parallelograms with different shapes and sizes. Thus, for each frame, each region estimates the probability of belonging to the foreground or background, to finally compute the consensus foreground mask among all the tilings. With this proposal, it is possible to detect the foreground in high resolution sequences, a process that is not feasible using pixel-level techniques. Several experiments have been carried out by employing a wide range of videos. A qualitative and quantitative comparison with the state-of-the-art algorithms is performed by using a well-known video dataset benchmark. The results show the feasibility of our proposal compared with higher resolution methods.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
In medical imaging, the lack of high-quality images is present in many areas such as magnetic resonance (MR). Due to many acquisition impediments, the generated images have not enough resolution to ...carry out an adequate diagnosis. Image super-resolution (SR) is an ill-posed problem that tries to infer information from the image to enhance its resolution. Nowadays, deep learning techniques have become a powerful tool to extract features from images and infer new information. In MR, most of the recent works are based on the minimization of the errors between the input and the output images based on the Euclidean norm. This work presents a new methodology to perform three-dimensional SR based on the combination of Lp-norms in the loss layer. Two multiobjective optimization techniques are used to combine two cost functions. The proposed loss layers were trained with the SRCNN3D and DCSRN networks and tested with two MR structural T1-weighted datasets, and then compared with the traditional Euclidean loss. Experimental results show significant differences in terms of Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM) and Bhattacharyya Coefficient (BC), while the residual images show refined details.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
The automatic detection and classification of vehicles in traffic sequences is a typical task which is carried out in many practical video surveillance systems. The advent of deep learning has ...facilitated the design of these systems. However, limitations in the resolution of the surveillance cameras imply that the vehicles are not clearly defined in the incoming video frames, which hampers the classification performance of deep learning Convolutional Neural Networks. In this paper a method is presented to overcome this challenge, which is based on several steps. An initial segmentation is followed by a postprocessing of the segmented images to solve vehicle overlapping and differing vehicle sizes. Then, a super resolution algorithm is employed to improve the definition of the image windows to be supplied to the neural networks. Finally, the outputs of an ensemble of such networks is integrated in order to obtain an improved recognition performance by the consensus of the networks of the ensemble. Several computational tests using well-known benchmarks demonstrate the effectiveness of the proposal, even in hard situations. Therefore, our vehicle classification system overcomes many limitations of naive application of Convolutional Neural Networks, since each proposed subsystem tackles different difficulties which arise in real traffic video data.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, UILJ, UKNU, UL, UM, UPUK
•Pan-Tilt-Zoom cameras allow viewing larger scene instead of using several fixed ones.•Automatically anomalous object detection is needed due to the large amount of videos.•Objects which are in the ...scene are detected by using a deep neural networks.•A Dirichlet-distribution algorithm determines which object is likely to be anomalous.•Most likely to be anomalous object is automatically tracked by PTZ camera controller.
Due to the large amount of visual information generated daily, proposals that automatically analyze and process data are becoming increasingly necessary. This work focuses on the detection of anomalous objects in video sequences captured by PTZ (pan-tilt-zoom) cameras, considering as anomalies the objects which belong to categories that should not appear in a specific scene (e.g. pedestrians on a highway). There is a lack in the previous literature of a principled approach for the control of PTZ cameras that takes advantage of the recent developments in deep learning-based object detection. Our proposal aims to fill this gap by offering a probabilistic framework where the guidance of PTZ cameras is accommodated. The proposed methodology involves three different modules. An object detection stage, where deep learning networks are used to detect the objects that appear in the scene; an anomalous detection module, where a mixture of Dirichlet distributions is considered to detect automatically, and without supervised training, those detected objects which are likely to be anomalous; and finally, a PTZ camera controller which allows to follow and focus on the object considered as the most probably anomalous in the scene. The experimental results show the performance and viability of our proposal, which outperforms several competitors from qualitative and quantitative points of view.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP