作者: Christian Kollmitzer
DOI: 10.1007/978-3-642-30721-8_16
关键词: Computer science 、 Subtraction 、 Computer vision 、 Object detection 、 Background subtraction 、 Computer stereo vision 、 Stereo cameras 、 Artificial intelligence 、 Stereo camera 、 Stereopsis 、 Object-class detection
摘要: This paper presents an improved method for detecting objects in stereo images and of calculating the distance, size speed these real time. can be achieved by applying a standard background subtraction on left right image, subsequently known as calculates disparity detected objects. calculation is supported several additional parameters like center object, color distribution object size. The used to verify plausibility calculate distance position this object. Out extracted, additionally calculated when tracked over frames. A dense map produced during learning phase serves possibility improve detection accuracy reliability.