Vol. 2 No. 2 (2023)

Eye diseases are one of the serious health problems affecting human life. Cataract, diabetic retinopathy and glaucoma eye diseases cause visual impairment and cause irreversible eye defects. Throughout human life; genetic, age and environmental factors affect people's eye health. Detection of the disease plays a critical role in order to apply the right diagnosis and therefore increase the quality of life of the patient. With the developing technology, artificial intelligence can detect eye defects and therefore whether there is a disease or not. This study aims to develop solutions for detecting an important health problem such as eye health by using deep learning models. In the related study, Convolutional Neural Networks models, one of the deep learning types are used. The data set used for disease detection includes a total of 2748 Retinal Fundus images taken from 1374 normal individuals and 1374 different disease groups. In order to compare the classification performances and to achieve better performance, a solution to the disease detection problem was sought by using a total of 5 different Convolutional Neural Networks architectures. These are DenseNet, EfficientNet, Xception, VGG, Resnet. For the validity of the approach, it was tested using the 10-fold cross-validation technique. Accuracy, Recall, Precision, F1-Score, and Matthews’s coefficient correlation metrics were used as performance evaluation criteria. When the classification performances were examined, the results obtained with the EfficientNet architecture were measured as 94.88%, 94.88%, 95.02%, 94.88%, 89.89% for Accuracy, Recall, Precision, F1-Score, and Matthews’s coefficient correlation metrics. In this context, the best classification performance was obtained with the EfficientNet architecture.

The use of welding technologies in the manufacturing sector plays a very important role and increases its popularity thanks to developing technologies. Welding technologies are used in almost every field where production takes place, and the speed and efficiency of welding technologies have increased in these sectors in recent years. The fact that artificial intelligence techniques are at the forefront and the efficient use of these techniques together with sensors has led to development in welding technology. Thus, welding robots emerged with the support of robots with artificial intelligence techniques, and adaptive systems that can adapt to different types of workpieces working autonomously in the manufacturing sector are shaping the sector. Despite these developments, non-autonomous systems are still used today by teaching the welding points to the robots by the operators. Along with the concept robotic system to be designed and implemented within the scope of this study, it is planned to determine the welding trajectory autonomously with artificial intelligence techniques, and to perform the welding process by following the welding trajectory by the robot.

Studies have been conducted on the application of Augmented Reality to support rehabilitation of motor function recovery. The goal of these studies is to facilitate functional recovery training through patient interaction with virtual objects generated by AR. Many of them use special devices such as depth sensors to superimpose virtual objects at appropriate positions in images, but a simple method that does not require such a device is desired. In order to realize superimposition using only a personal computer (PC) with a camera, this study utilizes a deep neural network that estimates the 3-dimensional (3D) coordinates of keypoints, such as human joints, from camera images. Specifically, a coordinate transformation matrix for superimposition is calculated from the 3D coordinates of keypoints. In order to clarify the effectiveness of this method, we conducted an experiment to evaluate the superimposition accuracy. The results show that the accuracy was highest in the space near the keypoints that had been used to compute the coordinate transformation matrix, and the accuracy was even higher when the number of keypoints was small. This indicates that this method is more suitable for localized training such as hand rehabilitation than for whole-body training. Since this method can be used only with a PC with a camera, it is expected to be widely used for rehabilitation support.

A form of shelled nut in the Betulaceae family is the hazelnut. The majority of it is grown in Türkiye internationally. It grows in the provinces of Türkiye's Black Sea region, which is a significant global production hub. Hazelnuts can be eaten in a variety of ways and are a great source of protein, fat, fiber, vitamins, and minerals. There are numerous applications for hazelnuts in the food business. This study uses pre-trained networks to categorize eight of the most popular hazelnut kinds farmed in Türkiye. In this study, locally named hazelnut varieties grown in Türkiye were examined. An automated computer vision system was used to capture the images of the different hazelnut kinds. Our dataset includes a total of 2722 images, consisting of 155 palaz, 340 yagli, 399 deve disi, 236 tombul, 399 damat, 354 cakildak, 437 kara findik, and 402 sivri hazelnuts. Using transfer learning, the DenseNet121 and InceptionV3 models of convolutional neural networks were employed to categorize these images. The dataset was split into training and testing portions, respectively. With InceptionV3 and DenseNet121, respectively, the research revealed classification accuracy of 96.99% and 96.18%.