Yusuf Sinan Akgül

Gebze Technical University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Ayşe Betül Oktay

Yıldız Technical University

Chandra Kambhamettu

University of Delaware

Nur Banu Albayrak

Kocaeli Üniversitesi

Abdullah Akay

Gebze Technical University

Ulaş Vural

Kocaeli Sağlık ve Teknoloji Üniversitesi

Sema Candemir

The Ohio State University

Ayse Serbetci

Istanbul Commerce University

Ilktan Ar

Kadir Has University

Maureen Stone

University of Maryland, Baltimore

Yakup Genç

Gebze Technical University

Cooperative Institutions

Gebze Technical University

Tübitak National Metrology Institute

Obafemi Awolowo University

Obafemi Awolowo University Teaching Hospitals Complex

Columbia University Irving Medical Center

University of Pennsylvania

Yıldız Technical University

Columbia University

Istanbul Medipol University

University of Botswana

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Planogram matching control in grocery products by image processing

Ünsal Gökdağ Mehmet Yasin Akpınar Yusuf Sinan Akgül

The placement of products in grocery shelves is designed according to several marketing strategies. These designs are represented by schemas named as planograms. Currently, the planogram matching control of grocery products are counducted by a personel manually and causes time loss. In this paper, a system that aims to solve this problem is proposed. The sytem takes a grocery shelf image as the input and gives the planogram compatibility as the output. Proposed system performs with %99.03 precision in shelf line detection, %97.31 precision product in detection and %99.21 accuracy in brand identification.

Grocery store

Product line

Image Matching

Identification

10.1109/siu.2016.7495972

Cite

Citations (3)

Adaptive Regularization Parameter for Graph Cut Segmentation

Lecture notes in computer science (2010)

Sema Candemir Yusuf Sinan Akgül

Regularization

Cut

Minification

10.1007/978-3-642-13772-3_13

Cite

Citations (25)

Stereo depth estimation using synchronous optimization with segment based regularization

Pattern Recognition Letters (2010)

Tarkan Aydın Yusuf Sinan Akgül

Regularization

Classification of discontinuities

Maxima and minima

10.1016/j.patrec.2010.07.012

Cite

Citations (9)

Ship positioning by matching radar images and map data

Ali Esref Aytac Orhan Aksoy Yusuf Sinan Akgül

Satellite network based GPS system is the most accurate, accountable and practical method for determining geographical ship locations. However, GPS systems are not preferable since they are prone to external interferences, (ie, jammers). In this study, a method for determining ship positions by matching navigation or surveillance radar images and satellite images is introduced. For this purpose, an estimate for radar image is generated by a novel method using satellite or map image. Then, the estimated radar image is compared with the real radar image using cross correlation to generate a success rate. Then, this success rate is improved iteratively by updating estimated ship position. The algorithm was run with both simulated radar images and real radar images to evaluate the correctness of the estimated ship positions. The study results are promising in the sense that introduced algorithm can be industrialized after improvements.

Man-portable radar

Radar lock-on

10.1109/siu.2014.6830506

Cite

Citations (1)

Fast non-linear video synopsis

Alparslan Yildiz Adem Özgür Yusuf Sinan Akgül

This paper presents a very efficient method for extracting the non-linear summary of a video sequence by synthesizing new summary frames from a number of original frames. The non-linear summary of a video sequence is fundamentally different from the classical video summarizing techniques which discard full frames. The high efficiency of the method is due to the employment of dynamic programming to find the space-time surfaces in the video volume that have lesser motion information. We project the video volume onto a plane orthogonal to one of its axes and find a minimum cost path on the time derivative of the projected image. Back-projecting the minimum cost path down to the video volume gives the space time surface to be discarded. Applying this process several times results in the summarized video. One of the major contributions of our method is that it can work on real-time video sequences with a small memory requirement. We demonstrated our method on several examples which are available at http://vision.gyte.edu.tr/projects. php?id=5.

Sequence (biology)

10.1109/iscis.2008.4717951

Cite

Citations (17)

Echocardiographic contour extraction with local and global priors through boosting and level sets

IEEE Computer Society Conference on Computer Vision and Pattern Recognition workshops (2009)

Ayşe Betül Oktay Yusuf Sinan Akgül

Extracting endocardium and epicardium from echocardiographic images is a challenging task because of large amounts of noise, signal drop-out, unrelated structures, and unseen wall parts. This paper introduces a new technique that automatically extracts cardiac borders by incorporating local and global priors through boosting and level set methods. The shape-based global prior is incorporated into the system by regularly re-initializing the level set surface under the influence of the expert detected contours. The local priors with image and temporal information are learned through boosting. The proposed system has many advantages. First, boosting encodes the knowledge about the image information and the temporal cardiac wall motion effectively by using spatiotemporal filters. Second, the local priors can use any features from the images including different filters and intensity profiles. Furthermore, other hard constraints like local shape, texture, distance, etc. can be added to local features effectively. The system is validated on echocardiograms and the results are found to be promising.

Boosting

Initialization

Prior information

10.1109/cvprw.2009.5204349

Cite

Citations (4)

Tree-Structured Binary Neural Networks

2022 30th Signal Processing and Communications Applications Conference (SIU) (2021)

Ayse Serbetci Yusuf Sinan Akgül

Deep Neural Networks are resource-intensive learning models. To enable their training and deployment on low-end devices, decreasing their computational requirements has become an attractive research area in recent years. One of the most promising approaches in this field is designing networks which use only binary values for the network weights and activations, i.e., Binary Neural Networks (BNN). However, BNNs have robustness and stability issues, which make their training difficult and BNNs perform considerably worse than their real-valued counterparts. To improve the training and deployment performance of BNNs, in this work, we employ a tree-structured network model, which has been widely used in Deep Neural Networks (DNN). We perform experiments to analyze the accuracy, stability and robustness of the proposed tree-structured BNN model. Tree-structured network model integrates multiple classifiers on a backbone BNN resulting in a multi-branch architecture. Our experiments demonstrates that tree-structures model improves the accuracy, stability and robustness of BNNs.

Robustness

Tree (set theory)

10.1109/siu53274.2021.9477811

Cite

Citations (0)

End-to-End Training of CNN Ensembles for Person Re-Identification

arXiv (Cornell University) (2020)

Ayse Serbetci Yusuf Sinan Akgül

We propose an end-to-end ensemble method for person re-identification (ReID) to address the problem of overfitting in discriminative models. These models are known to converge easily, but they are biased to the training data in general and may produce a high model variance, which is known as overfitting. The ReID task is more prone to this problem due to the large discrepancy between training and test distributions. To address this problem, our proposed ensemble learning framework produces several diverse and accurate base learners in a single DenseNet. Since most of the costly dense blocks are shared, our method is computationally efficient, which makes it favorable compared to the conventional ensemble models. Experiments on several benchmark datasets demonstrate that our method achieves state-of-the-art results. Noticeable performance improvements, especially on relatively small datasets, indicate that the proposed method deals with the overfitting problem effectively.

Overfitting

Discriminative model

Benchmark (surveying)

Ensemble Learning

Identification

Ensemble forecasting

Training set

10.48550/arxiv.2010.01342

Cite

Citations (0)

Estimation of the Prostate Volume from Abdominal Ultrasound Images by Image-Patch Voting

Applied Sciences (2022)

Nur Banu Albayrak Yusuf Sinan Akgül

Estimation of the prostate volume with ultrasound offers many advantages such as portability, low cost, harmlessness, and suitability for real-time operation. Abdominal Ultrasound (AUS) is a practical procedure that deserves more attention in automated prostate-volume-estimation studies. As the experts usually consider automatic end-to-end volume-estimation procedures as non-transparent and uninterpretable systems, we proposed an expert-in-the-loop automatic system that follows the classical prostate-volume-estimation procedures. Our system directly estimates the diameter parameters of the standard ellipsoid formula to produce the prostate volume. To obtain the diameters, our system detects four diameter endpoints from the transverse and two diameter endpoints from the sagittal AUS images as defined by the classical procedure. These endpoints are estimated using a new image-patch voting method to address characteristic problems of AUS images. We formed a novel prostate AUS data set from 305 patients with both transverse and sagittal planes. The data set includes MRI images for 75 of these patients. At least one expert manually marked all the data. Extensive experiments performed on this data set showed that the proposed system results ranged among experts’ volume estimations, and our system can be used in clinical practice.

Data set

Initialization

Software portability

10.3390/app12031390

Cite

Citations (10)

A Scale-Space Based Approach for Deformable Contour Optimization

Lecture notes in computer science (1999)

Yusuf Sinan Akgül Chandra Kambhamettu

Active contour model

Tracking (education)

Representation

10.1007/3-540-48236-9_36

Cite

Citations (7)