Color Features Integrated With Line Detection for Object Extraction and Recognition in Traffic Images Retrieval
This paper proposes a simple object extraction and recognition method with efficient searching for identifying and extracting the objects in a complex scene based on the color features. The background of the images is needed to be extracted and recognized in order to get the object of the interested in the images first. This can be achieved by getting the best separation line between building and road, followed by the interested objects (vehicles) on the road. The vehicle objects are represented by using Minimum Bound Rectangle (MBR) and the vehicle object representative points will be the left bottom coordinate of the MBR.