In order to improve the speed and accuracy of printing mesh fabric defect detection, a method based on Fourier transform and improved threshold segmentation is proposed. Taking the unpatterned printing mesh fabric as the research object, the Fourier transform is used for the collected printing mesh fabric images to acquire the spectral image of the printing mesh fabric images. Then the Butterworth low pass filter is used to the spectral image, and the highfrequency component of the Fourier space is filtered. Finally, the improved iterative threshold segmentation is used to the filtered images which are obtained by the inverse transform, so as to realize the detection of printing mesh fabric defect. The experimental results show that the algorithm can achieve the detection of dirt, scratches, fly wire, holes and other common defects quickly and accurately.