Tracking Targets in Sea Surface with the WiSARD Weightless Neural Network

Abstract

This paper presents a method of tracking sea surface targets in video using the WiSARD weightless neural network. The tracking of objects in video is an important and challenging task in many applications. Difficulties can arise due to weather conditions, target trajectory and appearance, occlusions, lighting conditions and noise. Tracking is a high-level application and requires the object location frame by frame in real time. At each frame, a tracker based on detection by segmentation performs three main steps: detection, tracking and analysis of the object characteristics. These steps depend on the segmentation quality and the tracking performed by the WiSARD neural network depends on the image binarization quality. This paper proposes a fast hybrid binarization (thresholding and edge detection) in YcbCr color model and ways to configure a WiSARD neural network to improve efficiency when binarization errors occur.