A5004972454- Advanced Image Processing Techniques
- Image and Signal Denoising Methods
- Image Enhancement Techniques
- Advanced Neural Network Applications
- Adversarial Robustness in Machine Learning
- Industrial Vision Systems and Defect Detection
- Advanced Vision and Imaging
- Anomaly Detection Techniques and Applications
- Advanced Image Fusion Techniques
Universidade Federal do Rio Grande
2019-2022
Convolutional Neural Networks stand the current state-of-the-art in image recognition, as well many computer vision tasks. Nevertheless, these architectures have been shown to be vulnerable manipulations, which may undermine reliability and safety of CNN-based models autonomous robotic applications. We present a rigorous evaluation robustness several high-level recognition investigate their performance under distinct distortions. propose testing framework emulates ill exposure conditions,...
Image restoration and image enhancement are critical processing tasks since good quality is mandatory for many applications. We particularly interested in the of ill-exposed images. These effects caused by sensor limitation or optical arrangement. They prevent details scene from being adequately represented captured image. proposed a deep neural network model due to number uncontrolled variables that impact acquisition. The can converge representative training data, loss, optimization...
Digital cameras work through transforming the scene's radiance into an electrical charge. Optical arrangement, sensors, and embedded electronics often limit accuracy of representation. Scenes with a dynamic range above capability camera or poor lighting are challenging conditions, which usually result in low contrast images. Soft clipping is compensated by power shifting image's histogram. However, under extreme ill exposure results severe that requires interpolation painting. We introduce...
The present work presents an artificial neural network architecture for the restoration of images damaged by underexposure and overexposure. problem is relevant in computer vision applications that are applied conditions where limitation sensor prevent scene details from being adequately represented captured image. This research attention-based composed two convolutional networks, one performs a preprocessing input image, while other enhancement degraded Regarding evaluation results, broad...
Image enhancement is a critical process in imagebased systems. In these systems, image quality crucial factor to achieve good performance. Scenes with dynamic range above the capability of camera or poor lighting are challenging conditions, which usually result low contrast images, and, that, we can have underexposure and/or overexposure problem. this work, our aim restore illexposed images. For purpose, present UCAN, small and fast learning-based model capable enhance poorly exposed The...