https://elcvia.cvc.uab.cat/issue/feed ELCVIA Electronic Letters on Computer Vision and Image Analysis 2022-09-13T07:29:30+02:00 Electronic Letters on Computer Vision and Image Analysis elcvia@cvc.uab.cat Open Journal Systems Electronic Journal on Computer Vision and Image Analysis https://elcvia.cvc.uab.cat/article/view/1465 Cricket Video Highlight Generation Methods: A Review 2022-07-07T14:57:21+02:00 Hansa Shingrakhia hansa.shingrakhia.2013@ieee.org Hetal Patel ec.hetal.patel@adit.ac.in <p>The key events extraction from a video for the best<br>representation of its contents is known as video summarization.<br>In this study, the game of cricket is specifically considered<br>for extracting important events such as boundaries, sixes and<br>wickets. The cricket video highlight generation frameworks<br>require extensive key event identification. These key events can<br>be identified by extracting the audio, visual and textual features<br>from any cricket video.The prediction accuracy of the cricket<br>video summarization mainly depends on the game rules, player’s<br>form, their skill, and different natural conditions. This paper<br>provides a complete survey of latest research in cricket video<br>summarization methods. It includes the quantitative evaluation<br>of the outcomes of the existing frameworks. This extensive review<br>highly recommended developing deep learning-assisted video<br>summarization approaches for cricket video due to their more<br>representative feature extraction and classification capability<br>than the conventional edge, texture features, and classifiers. The<br>scope of this analysis also includes future visions and research<br>opportunities in cricket highlight generation.</p> 2022-09-13T00:00:00+02:00 Copyright (c) 2022 Hansa Shingrakhia, Dr.Hetal Patel https://elcvia.cvc.uab.cat/article/view/1413 Diabetic foot ulcer segmentation using logistic regression, DBSCAN clustering and mathematical morphology operators 2022-07-07T14:58:50+02:00 Armando Heras-Tang armando.heras@matcom.uh.cu Damian Valdes-Santiago dvs89cs@gmail.com Ángela Mireya León-Mecías angela@matcom.uh.cu Marta Lourdes Baguer Díaz-Romañach mbaguer@matcom.uh.cu José Alejandro Mesejo-Chiong mesejo@matcom.uh.cu Carlos Cabal-Mirabal carlos.cabal@fisica.uh.cu Digital images are used for evaluation and diagnosis of a diabetic foot ulcer. Selecting the wound region (segmentation) in an image is a preliminary step for subsequent analysis. Most of the time, manual segmentation isn't very reliable because specialists could have different opinions over the ulcer border. This fact encourages researchers to find and test different automatic segmentation techniques. This paper presents a computer-aided ulcer region segmentation algorithm for diabetic foot images. The proposed algorithm has two stages: ulcer region segmentation, and post-processing of segmentation results. For the first stage, a trained machine learning model was selected to classify pixels inside the ulcer's region, after a comparison of five learning models. Exhaustive experiments have been performed with our own annotated dataset from images of Cuban patients. The second stage is needed because of the presence of some misclassified pixels. To solve this, we applied the DBSCAN clustering algorithm, together with dilation, and closing morphological operators. The best-trained model after the post-processing stage was the logistic regressor (Jaccard Index $0.81$, accuracy $0.94$, recall $0.86$, precision $0.91$, and F1 score $0.88$). The trained model was sensitive to irrelevant objects in the scene, but the patient foot. Physicians found these results promising to measure the lesion area and to follow-up the ulcer healing process over treatments, reducing errors. 2022-09-13T00:00:00+02:00 Copyright (c) 2022 Armando Heras-Tang, Damian Valdes-Santiago, Ángela Mireya León-Mecías, Marta Lourdes Baguer Díaz-Romañach, José Alejandro Mesejo-Chiong, Carlos Cabal-Mirabal https://elcvia.cvc.uab.cat/article/view/1538 Robust Pedestrian Detection and Path Prediction using Improved YOLOv5 2022-09-02T14:40:21+02:00 Kamal Omprakash Hajari kamalhajari123@gmail.com Ujwalla Gawande ujwallgawande@yahoo.co.in Yogesh Golhar ygolhar@stvincentngp.edu.in <pre style="-qt-block-indent: 0; text-indent: 0px; margin: 0px;"><span style="color: #000000;">In vision-based surveillance systems, pedestrian recognition and path prediction are critical concerns. Advanced computer vision applications, on the other hand, confront numerous challenges<br>due to differences in pedestrian postures and scales, backdrops, and occlusion. To tackle these challenges, we present a </span><span style="text-decoration: underline; color: #000000;">YOLOv5</span><span style="color: #000000;">-based deep learning-based pedestrian recognition and path prediction method. <br>The updated </span><span style="text-decoration: underline; color: #000000;">YOLOv5</span><span style="color: #000000;"> model was first used to detect pedestrians of various sizes and proportions. The proposed path prediction method is then used to estimate the pedestrian's path based on motion data. <br>The suggested method deals with partial occlusion circumstances to reduce object occlusion-induced progression and loss, and links recognition results with motion attributes. <br>After then, the path prediction algorithm uses motion and directional data to estimate the pedestrian movement's direction. The proposed method outperforms the existing methods, <br>according to the results of the experiments. Finally, we come to a conclusion and look into future study.</span></pre> 2022-09-13T00:00:00+02:00 Copyright (c) 2022 Kamal Omprakash Hajari, Dr. Ujwalla Gawande, Prof. Yogesh Golhar