Progressive edge extraction using multistage predictive coding

Autor: Yaqi Cheng, J. Wang, Chein-I Chang, Mark L.G. Althouse, Mann-Li Chang
Rok vydání: 2002
Předmět:
Zdroj: Proceedings of ICSIPNN '94. International Conference on Speech, Image Processing and Neural Networks.
DOI: 10.1109/sipnn.1994.344966
Popis: Presents a new image analysis technique for edge detection using multistage predictive coding (MPC). MPC is a progressive data compression technique which decomposes an image into a set of image components in multiple stages from which the original image can be recovered. The proposed coding scheme to implement MPC is multistage delta modulation (MDM) which includes a multistage quantizer for image decomposition and a priority code for image reconstruction. The utilization of the multistage quantizer is to decompose an image into multiple stages progressively in accordance with the significance of the image description. The task of the priority code is to prioritize all stages generated by the multistage quantizer; the lower the stage, the higher the priority. Using the priority code enables to extract features from the most significant edges to the least important details based on decreasing priorities. The experimental results are very encouraging and show that MDM is indeed a promising edge detection and feature extraction technique for real-time processing. >
Databáze: OpenAIRE