To reduce the computational load of Census transform, wavelet transform is used to down-sample the image. To solve the problem of the traditional Census transform overly relying on the central pixel and to improve the noise resistance, the weighted average grey value of the pixels in the support window is used as the reference value to construct the bit string instead of the grey value of the central pixel. A large number of scholars have made relevant improvements for the Census transform. In the local stereo matching algorithm, the Census transform used in the matching cost computation stage is more resistant and robust to illumination distortion, which is suitable for complex working environments, but the matching results for weak-textured regions are slightly worse and overly depend on the central pixel. The local stereo matching algorithm outperforms the global stereo matching algorithm in terms of real-time performance, but its matching accuracy is reduced. Binocular stereo matching algorithms can be classified into global stereo matching algorithms and local stereo matching algorithms. Binocular stereo matching is a crucial step in acquiring 3D information. With the Industry 4.0 process, binocular vision is becoming one of the main tools for obtaining three-dimensional (3D) information, such as height estimation and defect analysis. Because the construction of the adaptive region is based on strict criteria and comprehensive consideration, the algorithm proposed in the paper can improve the matching accuracy in the weak-textured regions and discontinuous disparity regions. The mismatching rate of this algorithm is greatly reduced, and the anti-noise performance is also improved considerably. Performance evaluations on Middlebury stereo data sets demonstrate that the proposed algorithm outperforms other seven most challenging stereo matching algorithms. Finally, the union of the two adaptive regions is used as the final support window of Census transform. In the cost computation stage, first, according to the preset arm length and colour threshold, a cross-based arm is obtained, which centres on the pixel to be matched then the adaptive regions of the vertical arm and the horizontal arm are constructed respectively, which have different shape and size. This paper proposes a stereo matching algorithm based on an improved adaptive support window. In binocular stereo matching, there has been a problem of low matching accuracy and noise immunity in discontinuous regions and weak-textured regions. IET Generation, Transmission & Distribution.IET Electrical Systems in Transportation.IET Cyber-Physical Systems: Theory & Applications.IET Collaborative Intelligent Manufacturing.CAAI Transactions on Intelligence Technology.

Mismatching device and image