Stereo-MAS: Multi-Agent System for Image Stereo Processing

Abstract

Abstract. This article presents a distributed agent-based architecture that can process the visual information obtained by stereoscopic cameras. The system is embedded within a global project whose objective is to develop an intelligent environment for location and identification within dependent environments that merge with other types of technologies. Vision algorithms are very costly and take a lot of time to respond, which is highly inconvenient if we consider that many applications can require action to be taken in real time. An agent architecture can automate the process of analyzing images obtained by cameras, and optimize the procedure.

People detection and stereoscopic analysis using MAS

“…The use of deliberative BDI (Belief, Desire, Intention) agents [3] [15] is essential in the development of the platform we are proposing. Apparently, the human visual system deals with a high level of specialization when it comes to classifying and processing the visual information that it receives, such as reconstructing an image by texture, shadow, depth, etc.…”

“…In addition to the specialized agents, there is an agent platform for monitoring and supervising the correct functioning of the system. A description and initial proposal for this architecture can be found in [15]. The system is comprised of a set of agents with defined roles that share information and services.…”

“…The data obtained from the stereoscopic camera are entered into the system and shared between the agents that will use specific services to process the data (filtering, preprocessing, disparity analysis, etc.) [15]. As shown in the figure, the system exits can be located in either the high-density disparity map that is obtained from the distance between the camera and the objects, the numerical representation of these distances, their three-dimensional representation in real time, and/or the detection of human forms in the specified area.…”

“…A commercial stereo camera [15] was employed in this work because it can capture two images from slightly different positions (stereo pair) that are transferred to the computer to calculate a disparity image containing the points matched in both images. Knowing the extrinsic and intrinsic parameters of the stereo camera it is possible to reconstruct the three-dimensional position.…”

“…Image processing applications are varied and include such aspects as remote control, the analysis of biomedical images, character recognition, virtual reality applications, and enhanced reality in collaborative systems, among others. Although image analysis and people detection is a well explored topic, the use of multiagent technology in this area has become the focal point of important interest [2] [15]. The availability of commercial hardware to solve the lowlevel problems of stereo processing has turned them into an attractive sensor to develop intelligent systems.…”