A rapid splicing method for vehicle bottom images

• Main Authors: GAO CHUNYAN, LYU SHENGFAN, LI PEIZHUO, ZHAO WENHUI
• Format: Patent
• Language: Chinese; English
• Published: 02.04.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention relates to a rapid splicing method for vehicle bottom images. The method is based on an SURF algorithm, retains the advantages of brightness resistance noise and scale change, when an image is inputted, reserves the low-frequency image with more basic information after wavelet transformation, and is characterized by after feature point extraction, establishing a main direction of thefeature point; calculating a difference value of main direction angles of feature points which may be correctly matched in the two images to be spliced and a difference value of corresponding scale factors, comparing the difference value with a correspondingly set threshold value, and removing the feature points again; reducing the dimension of the feature description vector through PCA principalcomponent analysis when the feature description vector is constructed, so that the time for extracting and matching the feature points is further shortened, the feature point matching pairs are obtained by using the ratio of