Road adjunct detection equipment, road adjunct detection method, and program

摘要

PROBLEM TO BE SOLVED: To automatically detect road appurtenances such as road signs and guard rails with high accuracy.SOLUTION: A storage device 6 previously stores image characteristic information 36 on the road appurtenances to be a detection object and form characteristic information 38 on a shape etc. of the detection object in object space. Image dividing means 20 divides an image which is photography of the object space into a plurality of partial image regions consisting of similar pixels respectively. Candidate image region extracting means 22 extracts a candidate image region corresponding to the detection object from among the partial image regions on the basis of the image characteristic information 36. Wall surface detecting means 24 detects wall surfaces of a feature on the basis of point group data indicating a three-dimensional shape of a surface of the feature in the object space. Candidate wall surface extracting means 26 extracts a candidate wall surface corresponding to the detection object from among the detected wall surfaces on the basis of the form characteristic information 38. Checking means 28 checks the candidate image region and a projection image of the candidate wall surface viewed from a photography location of the image, and detects the detection object on the basis of a checked result.