Browsing by Author "Özkanoğlu, Mehmet Akif"

Now showing 1 - 3 of 3

Open Access
InfraGAN: A GAN architecture to transfer visible images to infrared domain
(Elsevier BV * North-Holland, 2022-02-03) Özkanoğlu, Mehmet Akif; Ozer, S.
Utilizing both visible and infrared (IR) images in various deep learning based computer vision tasks has been a recent trend. Consequently, datasets having both visible and IR image pairs are desired in many applications. However, while large image datasets taken at the visible spectrum can be found in many domains, large IR-based datasets are not easily available in many domains. The lack of IR counterparts of the available visible image datasets limits existing deep algorithms to perform on IR images effectively. In this paper, to overcome with that challenge, we introduce a generative adversarial network (GAN) based solution and generate the IR equivalent of a given visible image by training our deep network to learn the relation between visible and IR modalities. In our proposed GAN architecture (InfraGAN), we introduce using structural similarity as an additional loss function. Furthermore, in our discriminator, we do not only consider the entire image being fake or real but also each pixel being fake or real. We evaluate our comparative results on three different datasets and report the state of the art results over five metrics when compared to Pix2Pix and ThermalGAN architectures from the literature. We report up to +16% better performance in Structural Similarity Index Measure (SSIM) over Pix2Pix and +8% better performance over ThermalGAN for VEDAI dataset. Further gains on different metrics and on different datasets are also reported in our experiments section.
Open Access
Object detection and synthetic infrared image generation for UAV-based aerial images
(2023-09) Özkanoğlu, Mehmet Akif
This thesis contains two main works related to aerial image processing. In the first work (in the first main part of this thesis), we present novel approaches to detect objects in aerial images. We introduce a novel object detection algorithm based on CenterNet which yields the state-of-the-art results in many metrics on many aerial benchmark datasets, when this thesis was written. In this part, we study the effect of different loss functions, and architectures for improving the detection performance of objects in aerial images taken by UAVs. We show that our proposed approaches help improving certain aspects of the learning process for detecting objects in aerial images. To train recent deep learning-based supervised object detection algorithms, the availability of annotations is essential. Many algorithms, today, use both infrared (IR) and visible (RGB) image pairs as input. However, large datasets (such as VisDrone [1] or ImageNet [2]) typically are captured in the visible spectrum. Therefore, a domain transfer-based approach to artificially generate in-frared equivalents of the visible images for existing datasets is presented in the second part of this thesis. Such image pairs, then, can be used to train object detection algorithms for either mode in future work.
Open Access
Offloading deep learning powered vision tasks from UAV to 5G edge server with denoising
(Institute of Electrical and Electronics Engineers, 2023-06-20) Özer, S.; İlhan, H. E.; Özkanoğlu, Mehmet Akif; Çırpan, H. A.
Offloading computationally heavy tasks from an unmanned aerial vehicle (UAV) to a remote server helps improve battery life and can help reduce resource requirements. Deep learning based state-of-the-art computer vision tasks, such as object segmentation and detection, are computationally heavy algorithms, requiring large memory and computing power. Many UAVs are using (pretrained) off-the-shelf versions of such algorithms. Offloading such power-hungry algorithms to a remote server could help UAVs save power significantly. However, deep learning based algorithms are susceptible to noise, and a wireless communication system, by its nature, introduces noise to the original signal. When the signal represents an image, noise affects the image. There has not been much work studying the effect of the noise introduced by the communication system on pretrained deep networks. In this work, we first analyze how reliable it is to offload deep learning based computer vision tasks (including both object segmentation and detection) by focusing on the effect of various parameters of a 5G wireless communication system on the transmitted image and demonstrate how the introduced noise of the used 5G system reduces the performance of the offloaded deep learning task. Then solutions are introduced to eliminate (or reduce) the negative effect of the noise. Proposed framework starts with introducing many classical techniques as alternative solutions, and then introduces a novel deep learning based solution to denoise the given noisy input image. The performance of various denoising algorithms on offloading both object segmentation and object detection tasks are compared. Our proposed deep transformer-based denoiser algorithm (NR-Net) yields state-of-the-art results in our experiments.