Sürdürülebilir Kalkınma Amaçları
Yayınlar
- BildiriYalnız künye2026Analysis of Temporal Consistency in Video-Based Palmprint ROI Tracking
Sunnetci, Betul Kosmaz; Bingol, Ozkan; Gedikli, Eyup
In real-time palmprint recognition systems, operating under unconstrained conditions, maintaining the defined palmprint region of interest (ROI) throughout hand motion presents a significant challenge. Variations in hand position or orientation may cause the ROI to shift unexpectedly or partially or completely disappear, which can adversely affect subsequent processing steps and their computational cost. In this study, the temporal stability and consistency of palmprint ROIs extracted from video data in the MPW-180 dataset were evaluated using two different approaches. Specifically, the behavior of two methods (optical flow based tracking and keypoint-based MediaPipe Hands detection) was quantitatively analyzed over time using corner position shifts and overlap ratios. Experimental results indicate that although the keypointbased method produces spatially accurate ROI estimates, it may exhibit sudden positional jumps in certain frames. In contrast, the optical flow method provides smoother motion transitions but may introduce positional drift over time due to the accumulation of tracking errors.
- MakaleAçık Erişim2024EEG Sinyallerinden Meta-Sezgisel Optimizasyon Algoritmalarına Dayalı Özellik Seçimi
Gedikli, Eyup; Yurdusever, Taner
Özellik seçimi makine öğrenmesi alanında, sınıflandırma işlemlerinin öncesinde bulunan önemli bir aşamadır. Ayırt edici özelliklerin iyi belirlendiği durumlarda, sınıflandırma başarı performası artar ve daha az özellik değerlendirildiği için hesaplama maliyeti azalır. Elektroensefalografi (EEG) yöntemi ile beynin spontan elektrik aktiviteleri ölçülmektedir. EEG sinyallerinin analiz edilmesiyle, duygu durumu, hastalık teşhisi, anomali tespiti yapılabilmektedir. Bu çalışmada, EEG sinyallerinden epilepsi teşhisi amacıyla, sınıflandırmada kullanılan özelliklerin seçilmesine çalışılmıştır. Kullanılan verisetinde, epileptik ve sağlıklı bireylerden oluşan 2 sınıf mevcuttur. Verisetinde, EEG sinyallerinin alt bileşenlerinden 667 özellik vardır. Sınıflandırma için bu 667 özelikten meta-segisel optimizasyon algoritmaları ile ayırt edici özellikler seçilmiştir. Sınıflandırma için k en yakın komşuluk algoritması kullanılmıştır. EEG sinyallerinin alt bileşenlerinin tamamı ile yapılan sınıflandırmada, %60,05 doğruluk başarısı elde edilmiştir. Gri Kurt Optimizasyonu, Balina Optimizasyonu ve Harris Şahinler Optimizasyonu metasezgisel algoritmaları ile özellik seçimi sonucunda, sınıflandırma başarısı %80,16 olarak elde edilmiştir. Bu sınıflandırma başarısı 5-10 özellik kullanılarak elde edilebilmektedir. Sonuç olarak meta-sezgisel optimizasyon algoritmaları ile daha az özellik seçilerek doğruluk oranı artmış ve hesaplama maliyeti azalmıştır.
- BildiriAçık Erişim2024Cross-view Gait Recognition Based on Fine-Tuned Deep Networks
Yaprak, Busranur; Gedikli, Eyup; Bingöl, Ozkan; Dogan, Ramazan Ozgur
Gait recognition is a biometrics-based computer vision process used to identify people based on their walking styles. Compared to other types of biometrics, gait offers a more advantageous recognition process as it does not require high-resolution and close-range images and obtains without contact. But besides this, gait biometrics is highly affected by cross-view variation, and under this variation recognition performance decreases significantly. In this study, performance evaluations of fine-tuned VGG-16 and ResNet-50 deep CNN networks on the cross-view gait recognition problem are performed. For this purpose, Gait energy images (GEI) and Silhouettes obtained from CASIA-B, the most comprehensive data set in gait recognition, are given as input to the networks. The experimental results showed that the VGG-16 network achieved higher recognition rates in cross-view gait recognition.
- BildiriAçık Erişim2024In Different Scenarios MobileNet-V1 for Cross-view Gait Recognition
Yaprak, Busranur; Gedikli, Eyup; Badica, C.; Ivanovic, M.; Koprinkova-Hristova, P.; Leon, F.; Manolopoulos, Y.; Yildirim, T.; Ucar, A.
Gait recognition aims to identify people from a distance by analyzing their walking style. Nevertheless, the efficacy of recognition drops significantly under cross-view and, appearance-based variations such as carrying and clothing. In this study, the performance of the MobileNet-V1 deep network is evaluated in various scenarios to address the cross-view gait recognition problem. In the first scenario, the fine-tuned MobileNet-V1 is evaluated on Gait Energy Images (GEI) as input data, while in the second scenario, the fine-tuned MobileNet-V1 is assessed with Optical Flows and masked RGB frames input data. In the last scenario, the first two scenarios are combined over a single fused deep network based on finetuned MobileNet-V1, and a single recognition process is performed using two different fused features data; GEI features with Optical Flow features, and GEI features with masked RGB frame features. In the evaluation process, a comprehensive data set for the cross-view gait recognition problem, CASIA-B is used for the experiments. The obtained results demonstrate that in the last scenario, the contribution of masked RGB frame features to the recognition rate of GEI is more significant.
- BildiriYayıncıda açık erişim2024High Performance Automatic ROI Extraction from Hand Images with Convolutional Neural Networks
Kosmaz Sunnetci, Betul; Bingol, Ozkan; Gedikli, Eyup; Ekinci, Murat
Automatic ROI extraction from hand images is a critical step in palmprint recognition. Traditional methods involve steps such as identifying the region of the hand, finding the valleys between the fingers and marking the palm area. Some of these processes are performed manually and some are performed with image processing techniques. However, these methods are time-consuming, vulnerable to human error, and computationally redundant. In this study, a deep learning-based convolutional neural network is developed for high-performance automatic ROI extraction from hand images. The proposed method eliminates the hand segmentation process and provides a solution where the palm region can be automatically selected directly. This reduces processing time, minimizes human error and improves overall accuracy.
- BildiriYalnız künye2019Meta-Heuristic Algorithms Based Multi-Level Thresholding
Küçükugurlu, Busranur; Gedikli, Eyup
Thresholding is a very important stage in computer vision applications. An ideal single thresholding algorithm is not available for all environments. Multi-level thresholding in environments with multiple objects is constantly being developed for interpretation of images. Kapur entropy and Otsu approaches are among the most successful algorithms in the literature. In this study, it is tried to increase the performance of Otsu and Kapur algorithms by using meta-heuristic optimization approaches. The results of the Firefly Algorithm (FF) and Real Coded Genetic Algorithm (RGA) were evaluated with PSNR, SSIM and CPU processing time criteria.