作者: Kosaka Akio
DOI:
关键词: Feature (computer vision) 、 Line (geometry) 、 Computer vision 、 Noise (video) 、 Component (UML) 、 Image processing 、 Artificial intelligence 、 Object (computer science) 、 Digital image 、 Signal 、 Computer science
摘要: PURPOSE:To provide an image recognizing device which can take optimum matching with the model of object given in advance even when a contour part is separately extracted by influence noise or shape affected deformation like. CONSTITUTION:This provided storage means 13 for storing feature as set components, pick up 11 picking and transmitting video signal, 12 converting signal into digital this image, processing 14 extracting straight line component curved corresponding to from dividing groups while considering connection relation component, transforming it taking between feature.