Objective：This study proposes a novel method to address the issues of low efficiency, low accuracy, and low automation prevalent in manual measurement of diameter level and wall thickness of raw bamboo. The method utilizes a pulsed line array laser camera and is based on the OpenCV image processing library and encoder principle. Methods：Firstly, the median filter is used to reduce the noise in the collected bamboo tube images; then, the global threshold segmentation method is used to separate the background and the texture information of the inner wall of the bamboo tube, and the cross-section profile of the bamboo tube is extracted by the Canny operator; finally, the diameter of the bamboo tube and the wall thickness dimensions of the bamboo tube are obtained by calculating based on the coordinates of the pixel points of the extracted profile. Results：The experimental results show that for 12 different sizes of bamboo tubes selected arbitrarily, the accuracy of this algorithm reaches 95.12% and 93.25% for the diameter and wall thickness measurement results, respectively, while the average measurement time is 810 ms, which is much better than manual measurement. Conclusion ：The machine vision-based detection method for moso bamboo′s diameter grade and wall thickness can satisfy required accuracy levels and enhance measurement efficiency, providing a theoretical foundation for the replacement of manual inspections with machine vision technology.