Abstract <p>Aiming at the problems of dense background noise, low contrast and blurred edges in the extraction of weak defects from aerospace organic glass, an innovative method combining two-dimensional (2D) adaptive gamma correction and a histogram bimodal threshold segmentation algorithm is proposed to enable accurate extraction of these defects. By constructing a two-dimensional adaptive gamma correction function to extract the brightness V channel and performing a multi-scale Gaussian convolution operation on the image, this method effectively suppresses background noise and significantly enhances image contrast. Additionally, a bimodal detection function is constructed to locate the extreme points of the histogram, and the optimal threshold is dynamically calculated to realize the binarization of the damaged area, while fully retaining weak defect details and mitigating edge blurring. The experimental results show that the proposed method can accurately extract the weak defects of aerospace organic glass, with a segmentation accuracy of 0.9992 and a precision of 0.9047, which effectively overcomes the difficulty of extracting blurred edges and significantly improves the reliability of weak defect extraction.</p>