Building expert systems
Model-based recognition in robot vision
ACM Computing Surveys (CSUR)
Programming in Prolog
VITS-A Vision System for Autonomous Land Vehicle Navigation
IEEE Transactions on Pattern Analysis and Machine Intelligence - Special Issue on Industrial Machine Vision and Computer Vision Technology:8MPart
Vision and navigation for the Carnegie-Mellon navlab
IEEE Transactions on Pattern Analysis and Machine Intelligence - Special Issue on Industrial Machine Vision and Computer Vision Technology:8MPart
Expert systems: knowledge, uncertainty, and decision
Expert systems: knowledge, uncertainty, and decision
Rule-Based Interpretation of Aerial Imagery
IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic segmentation and inpainting of specular highlights for endoscopic imaging
Journal on Image and Video Processing - Special issue on emerging methods for color image and video quality enhancement
Expressing relational and temporal knowledge in visual probabilistic networks
UAI'92 Proceedings of the Eighth international conference on Uncertainty in artificial intelligence
| Hi-index | 0.00 |
Endoscopy is a complex task in which an expert physician is required to guide the endoscope inside the human colon. The objective of this system is to develop a computer assistant that could help the doctor with the navigation of the endoscope inside the colon, serving as an advisory system for learning endoscopists.A knowledge-base (KB) in colon endoscopy has been compiled from the knowledge extracted from an expert colonoscopist. It includes knowledge for colon image interpretation and for control of the endoscope. Using features from intermediate vision we are using the expert rules to recognize the important objects in the images, in the first stage, and later for advise and control. In particular, we are interested in detecting “special” situations in the colon for which the expert heuristics are useful.An initial prototype has been implemented using a parallel architecture with transputers and a PC. The expert system is implemented in Prolog and it communicates with the feature extraction programs running in a “transputer pyramid” by transforming the vision features into symbolic predicates for logical inference. We have tested the system with real colon images from a videotape to identify the lumen. Work is in progress to extend the feature extraction process so more objects could be recognized by the system.