Research on data fusion distortion invariant target recognition technology in joint transform correlator

• Published in: 2009 International Conference on Mechatronics and Automation pp. 3955 - 3959
• Main Authors: Lu Yang, Huanhuan Jia, Cuiling Zhao, Wensheng Wang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2009
• Subjects: Computer simulation; Correlators; data fusion; distortion invariant; Fourier transforms; Image recognition; Image resolution; joint transform correlator(JTC); Lenses; minimum average correlation energy (MACE); Optical distortion; Optical filters; Optical sensors; Target recognition
• ISBN: 1424426928; 9781424426928
• ISSN: 2152-7431
• DOI: 10.1109/ICMA.2009.5244874

In optical correlation target auto-detection and recognition, distortion invariant (rotation invariant and scale invariant) image recognition is the hard technology. The main character of minimum average correlation energy (MACE) filters in image recognition is high peak acuity, but it is sensitive to distortion, so its application is limited. Combine data fusion technology with the algorithm of MACE, a new algorithm which is called data fusion distortion invariant joint transform correlate is presented. The problem of distortion invariant image recognition is resolved very well by this algorithm. Computer simulation and experimental results show that the problems of target recognition with rotation variation and scale variation can be resolved by using the data fusion distortion invariant algorithm in hybrid optoelectronic real-time joint transform correlator. The contrast of correlation peak is enhanced greatly, and the power of target result of the ground target-automobile is presented. It proves that the data fusion distortion invariant algorithm is feasible.