Papers by Kerim Kürşat Çevik
Springer International Publishing eBooks, 2023
Mühendislik Bilimleri ve Tasarım Dergisi
Gelişimsel kalça displazisi (GKD) terimi, femur başı ve asetabulumun yanlış hizada olduğu, anorma... more Gelişimsel kalça displazisi (GKD) terimi, femur başı ve asetabulumun yanlış hizada olduğu, anormal şekilde büyüdüğü veya her ikisinin birden olduğu yeni doğanları etkileyen bir dizi kalça anormalliği olarak tanımlanır. Ultrasonografik değerlendirme tekniği, ultrasonografi uzmanının kesin hesaplamalar için kullanılan doğru çerçeveyi(standart düzlem) seçme yeteneğine dayanır. Çalışmamızda, gerçek zamanlı 2B ultrason görüntülerinden standart düzlemi belirleyen ve her bir sonuç için doğruluk oranını hesaplayan yeni bir bilgisayar destekli sistemi geliştirilmiştir. Bu işlemler için literatürde son zamanlarda kullanılan derin öğrenme mimarilerinden yararlanılmıştır. Ayrıca önceden eğitilmiş ağlar (SqueezeNet, VGG16, VGG19, ResNet50 ve ResNet101) kullanılarak, sistemin performansını artırmak için transfer öğrenmesi gerçekleştirilmiştir. Nesne algılamanın en iyi yöntemlerinden biri olan You Only Look Once (YOLO) modeli, DDH konumunu belirlemek için önceden eğitilmiş ağlarla birlikte kullanı...
Veterinary Sciences
This study investigates the implementation of deep learning (DL) approaches to the fertile egg-re... more This study investigates the implementation of deep learning (DL) approaches to the fertile egg-recognition problem, based on incubator images. In this study, we aimed to classify chicken eggs according to both segmentation and fertility status with a Mask R-CNN-based approach. In this manner, images can be handled by a single DL model to successfully perform detection, classification and segmentation of fertile and infertile eggs. Two different test processes were used in this study. In the first test application, a data set containing five fertile eggs was used. In the second, testing was carried out on the data set containing 18 fertile eggs. For evaluating this study, we used AP, one of the most important metrics for evaluating object detection and segmentation models in computer vision. When the results obtained were examined, the optimum threshold value (IoU) value was determined as 0.7. According to the IoU of 0.7, it was observed that all fertile eggs in the incubator were de...
2016 24th Signal Processing and Communication Application Conference (SIU), 2016
In our study, Developmental Dysplasia of the Hip (DDH) is intended to automatically segmenting th... more In our study, Developmental Dysplasia of the Hip (DDH) is intended to automatically segmenting the ultrasound images for diagnosis. Initially, a filter is applied to the raw images. Seven different filters (Mean, Median, Gaussian, Wiener, Perona & Malik, Lee and Frost) are applied to the images and finally the output images are evaluated. Filtered DDH images were segmented and results are evaluated in the second part of the work. In the DDH diagnosis, the ilium and femoral regions are segmented by using Active Contour Models and Circular Hough Transform methods, respectively. When the segmentation process is analyzed, it is observed that the Wiener filters manage to increase the success rate due to their ability to remove speckle noise and ilium segmentation was performed with 94%. It is observed that Wiener filter was also success, besides when applied histogram equalization after filtering success rate is determined as 96% in the femoral region.
Iranian Journal of Radiology, 2016
Background: Developmental dysplasia of the hip (DDH) can be detected with ultrasonography (USG) i... more Background: Developmental dysplasia of the hip (DDH) can be detected with ultrasonography (USG) images. However, the accuracy of this method is dependent on the skill of the radiologist. Radiologists measure the hip joint angles without computer-based diagnostic systems. This causes mistakes in the diagnosis of DDH. Objectives: In this study, we aimed to automate segmentation of DDH ultrasound images in order to make it convenient for radiologic diagnosis by this recommended system. Materials and Methods: This experiment consisted of several steps, in which pure DDH and various noise-added images were formed. Then, seven different filters (mean, median, Gaussian, Wiener, Perona and Malik, Lee, and Frost) were applied to the images, and the output images were evaluated. The study initially evaluated the filter implementations on the pure DDH images. Then, three different noise functions, speckle, salt and pepper, and Gaussian, were applied to the images and the noisy images were filtered. In the last part, the peak signal to noise ratio (PSNR) and mean square error (MSE) values of the filtered images were evaluated. PSNR and MSE distortion measurements were applied to determine the image qualities of the original image and the output image. As a result, the differences in the results of different noise removal filters were observed. Results: The best results of PSNR values obtained in filtering were: Wiener (43.49), Perona and Malik (27.68), median (40.60) and Lee (35.35) for the noise functions of raw images, Gaussian noise added, salt and pepper noise added and speckle noise added images, respectively. After the segmentation process, it was seen that applying filtering to DDH USG images had low influence. We correctly segmented the ilium zone with the active contour model. Conclusion: Various filters are needed to improve the image quality. In this study, seven different filters were implemented and investigated on both noisy and noise-free images.
WOS: 000466018400018This research aimed to determine the fertilization control of the eggs in an ... more WOS: 000466018400018This research aimed to determine the fertilization control of the eggs in an incubator between 0th and 5th days by image processing techniques via low-priced tools. Three different datasets that were composed of eggs whose images taken at different times in the incubator were prepared. Several filtering and morphology methods, gray level conversion and dynamic thresholding were utilized to process the 15 egg images. Moreover, the original processing codes based on the problem were given. White and Black percentages of binary images were utilized to determine the egg control. According to the test results, for the first dataset; 73.34% of fertility accuracy was achieved on the third day; 100% of fertility accuracy was achieved on the fourth day, for the second dataset; 93.34% of fertility accuracy was achieved on the third day; 93.34% of fertility accuracy was achieved again on the fourth day; for the third dataset, 93.34% of fertility accuracy was achieved on the...
IEEE Access, 2020
The number of animals worldwide is increasing day by day to meet the increasing animal protein ne... more The number of animals worldwide is increasing day by day to meet the increasing animal protein needs. Depending on the increase in animal production, yield amount which can be obtained from per unit area can be increased by increasing the number of animals. In dairy cattle farms, it is necessary to group the animals according to their body condition score (BCS) and to care and feed the animals at certain times. Under normal conditions, these processes should be conducted by animal caregivers or experts coming to the enterprise. BCS ratings conducted by experts on farms based on visual examination may give unreliable results and may include misinterpretations. Therefore, technology-supported systems are required. In this study, the prediction of BCS, which is the most important indicator of proper feeding of dairy cattle, is aimed. In addition, by adapting the designed system to simple, fast and user-friendly mobile software, it will be possible to provide tests in enterprise environments in a shorter time. In the design of the system, deep learning models, which have been used frequently in recent years in computer science, have been used. The CNN model, which was trained with a 94.69% success rate through these data, has been converted into a mobile-friendly format for real-time tests. It is aimed to make real-time tests and provide easy access for dairy producers with the help of the designed mobile software. In order to increase the success of the CNN architecture, pre-trained networks have been utilized. In the study, VGG19 pre-trained network, whose success rate has been proved in the previous studies conducted in the literature, was used in model design. The 78.0% performance results obtained from the study indicate that pre-trained CNN architectures based on deep learning are successful for the real-time BCS classification problem.
2019 Innovations in Intelligent Systems and Applications Conference (ASYU), 2019
The most important indicator of whether animals 'needs are met in livestock enterprises is th... more The most important indicator of whether animals 'needs are met in livestock enterprises is the animals' body condition score (BCS) score. In dairy cattle BCS is based on scoring from 1 to 5 according to the external appearance of the animals. BCS is a subjective method based on visual or palpation method to determine the relationship between subcutaneous fat thickness and bone protrusions in pelvic region in back, waist and coccyx regions in cattle. Generally, BCS values in the enterprises are determined by a method based on expert knowledge and determined by observation. If the animal is above or below the desired BCS, at this stage, diseases resulting from metabolic problems, low yield or animal losses may be observed. With the regular control of this situation, the profitability of the enterprise may increase with the production of more health animals. For this purpose, it was aimed to determine the BCS score with a computer-aided software. Images from cattle were arrange...
Importance of road lighting has increased day by day to provide drivers to travel in safe and com... more Importance of road lighting has increased day by day to provide drivers to travel in safe and comfort as the result of increasing vehicle traffic. Ideal luminance values based on the type of road is specified in 115 numbered technical report of the International Commission on Illumination. For this technical report, there are illumination classifications for five different road types. M3 road lighting group is demanded for urban main routes (speed < 50 km/h); ideal luminance value for this group is accepted as 1 cd/m2 at least. There occur damages in time based on the use of lamps, environmental effects, and pollution factor; there also occurs decreases in total luminance. Photometric measurement of luminaires needs to be periodically performed to determine how much the luminaires are affected by these problems. Illumination measurements are made by photometric measuring instruments. Time, cost and qualified manpower are necessary for this process. Artificial intelligence-based m...
INTERNATIONAL JOURNAL OF INFORMATICS TECHNOLOGIES, 2012
Bu calismada, yapay sinir aglarinin egitiminde kullanilmak uzere C#.NET programlama dili ile gors... more Bu calismada, yapay sinir aglarinin egitiminde kullanilmak uzere C#.NET programlama dili ile gorsel arayuze sahip olan bir egitim yazilimi gelistirilmistir. Gerceklestirilen yapay sinir agi siniflandirma yazilimi rakam goruntulerinin siniflandirilmasi ornegine uygulanmis ve basarili sonuclar edilmistir. Ayrica yapay sinir aglari konusunun onemi anlatilmis ve ogrenebilen yazilimlarin bilgisayar teknolojilerindeki avantajlarina deginilmistir. Yapay sinir aglarinin egitilmesi asamasinda kullanilan mevcut yazilimlarin kullanim karmasikliginin ve zorlugunun onune gecmek icin gelistirilen yazilimda rakam orneklerinin taninmasinda geriye yayilim algoritmasindan yararlanilmistir. Yazilim ile olusturulan egitim ortami uzerinden sonuclar icin basari analizi yapilabilmekte ve performansi olculebilmektedir. Yazilim yapay sinir aglari ile cozulebilecek uygulamalarda ve akademik calismalarda kullanilacak yapida tasarlanmistir.
2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), 2019
In recent years, especially in industrial automation systems, in order for robots to understand t... more In recent years, especially in industrial automation systems, in order for robots to understand their distances and positions according to the target, close to human vision, computer vision systems are needed. Computer vision close to human vision can only be created using stereo cameras. In this study, a computer vision system is developed using the stereo camera system for measuring object distances. In the study, first of all, for the distance measurement, the distance of the face images obtained from the stereo camera system to the screen is calculated. In measuring the distances of the face images to screen, the disparity maps are first extracted and the face region is detected. Afterwards, the distance measurements are performed on the obtained images in the stereo camera system on account of calculating the shifts between the frames. In the experimental studies, the actual distance values such as 71, 74, 75, 79, 110, 125, and 115 of the face to the screen are measured as 70, 72, 73, 77, 97, 120, 132 cm by proposed system, respectively. When the experimental results are examined, we can say that the proposed computer vision system is successful in distance measurement using stereo camera view.
Handbook of Research on Disease Prediction Through Data Analytics and Machine Learning, 2021
This chapter aimed to evaluate heuristic approach performances for artificial neural networks (AN... more This chapter aimed to evaluate heuristic approach performances for artificial neural networks (ANN) training. For this purpose, software that can perform ANN training application was developed using four different algorithms. First of all, training system was developed via back propagation (BP) algorithm, which is the most commonly used method for ANN training in the literature. Then, in order to compare the performance of this method with the heuristic methods, software that performs ANN training with genetic algorithm (GA), particle swarm optimization (PSO), and artificial immunity (AI) methods were designed. These designed software programs were tested on the breast cancer dataset taken from UCI (University of California, Irvine) database. When the test results were evaluated, it was seen that the most important difference between heuristic algorithms and BP algorithm occurred during the training period. When the training-test durations and performance rates were examined, the op...
Interactive Learning Environments, 2019
ABSTRACTThe aim of this study is to demonstrate effectiveness of mobile game called SOS Table in ... more ABSTRACTThe aim of this study is to demonstrate effectiveness of mobile game called SOS Table in the context of the subject “Tenses in English” within the framework of mobile learning tools accepta...
Pakistan Journal of Zoology, 2020
AJIT-e: Online Academic Journal of Information Technology, 2018
Journal of Medical Imaging and Health Informatics, 2016
TURKISH JOURNAL OF VETERINARY AND ANIMAL SCIENCES, 2020
In this study, ration preparation software to minimize the cost of feed for ruminant livestock su... more In this study, ration preparation software to minimize the cost of feed for ruminant livestock such as cattle, sheep, and goats for both milk and meat yield was developed for Web-and Android-based systems using genetic algorithms. To maximize accessibility on PCs, smartphones, and tablet PCs, we used Web-and Android-based software to find cheaper feed mixes that satisfy the nutritional requirements of ruminants. With this novel system, farmers and scientists can obtain low-cost feed mixes via the Web or smartphones, regardless of time or location. This application is useful for feed producers and farmers because they can use this software from any location and at any time. Users can input their new feed resources for preparing rations.
Bu calismada, bir giris kapisina gelen aracin plakasinin goruntu isleme algoritmalari ile taninar... more Bu calismada, bir giris kapisina gelen aracin plakasinin goruntu isleme algoritmalari ile taninarak kapinin otomatik olarak acilip kapanmasini saglayan bir sistem gelistirilmistir. Calismada, insan gucunden tasarruf saglamak ve guvenligi artirmak amaclanmistir. Gelistirilen sistemde, kapiya gelen aracin goruntusu kamera ile alinarak bilgisayara aktarilmakta ve C# programlama dili kullanilarak gelistirilen goruntu isleme algoritmalari tabanli bir yazilim sayesinde plaka tanima islemi cok hizli bir sekilde gerceklestirilmektedir. Yazilim tarafindan okunan plaka veri tabanindan kontrol edilmekte ve plaka veri tabaninda kayitli ise kapi acilmakta, kayitli degilse uyari verilmektedir. Calismada plakadaki karakterleri ayirmak icin Blob Coloring (Ikili Renklendirme) algoritmasi kullanilmistir. Gelistirilen yazilim ile plakalarin yerinin bulunmasinda %98, plakalarin dogru olarak okunmasinda ise % 88,1 basari elde edilmis ve daha onceki calismalara oranla bir iyilestirme saglandigi gorulmustur.
International Journal of Applied Mathematics, Electronics and Computers, 2016
In this article, a study was carried on ultrasound (US) images for the automatic diagnosis of the... more In this article, a study was carried on ultrasound (US) images for the automatic diagnosis of the disease of the developmental hip dysplasia (DDH). It was aimed with this study at minimizing the errors of the experts in DDH diagnosis. As a first step in the study; commonly known as the images and reduce noise in the US image, image filter are applied to improve the quality. In the second stage; by using Active Contour Model method it was determined acetabular roof and labrum areas. In the third stage; alpha and beta angles that is necessary to be applied Graf method and used DDH diagnosis are determined by using various morphological image algorithms on the image. In the last stage, the classification of Graf method was made and the performance of the system was measured by comparing expert data and the results. According to type conditions of Graf method, in the images of 40 out of 50 it was found the same due to software which was designed with expert data. In the remaining 10 ima...
Uploads
Papers by Kerim Kürşat Çevik