Hazard Control in Industrial Environments: A Knowledge-Vision-Based Approach

Abstract

This paper proposes the integration of image processing techniques (such as image segmentation, feature extraction and selection) and a knowledge representation approach in a framework for the development of an automatic system able to identify, in real time, unsafe activities in industrial environments. In this framework, the visual information (feature extraction) acquired from video-camera images and other context based gathered data are represented as Set of Experience Knowledge Structure (SOEKS), a formal decision event for reasoning and risk evaluation. Then, grouped sets of decisions from the same category are stored as decisional experience Decisional DNA (DDNA) to support future decision making events in similar input images. Unlike the existing sensor and vision-based approaches, that required rewriting most of the code when a condition, situation or requirement changes, our platform is an adaptable system capable of working in a variety of video analysis scenarios. Depending on the safety requirements of each industrial environment, users can feed the system with flexible rules and in the end, the platform provides decision makers with hazard evaluations that reuse experience for event identification and correction.