Maximum Average Correlation Height (MACH) algorithm was applied on JTC to achieve distortion target recognition to solve the problem of inaccurate recognition for distorted target (the targets rotated in plane or scale changed) in cluttered background with Joint Transform Correlator (JTC). First of all, three performance control parameters of MACH filter were optimized in frequency domain, and then the optimized MACH frequency domain filter was projected to object space; the MACH reference template could be obtained. As practical examples, the optical experiments on warship and car in clutter background were presented. The results show that when the control parameters of the filter are γ=1, α=0 and β=0.1, the obtained MACH reference template is clearest, the correlation peak of the target is sharpest and brightest. In this way, the obtained scale distortion tolerance on the JTC is 28%, and rotation distortion tolerance is 20°. For the target which exceeding the distortion tolerance range, the correlation peak is weaker and weaker, but some can still be recognized. In the field of distorted target detection, the improved MACH filter has a powerful advantage and a good application prospect.