Nondestructive testing method for solid rocket engine based on machine learning and computer vision

The invention provides a nondestructive testing method for a solid rocket engine based on machine learning and computer vision. An input image is divided into a plurality of local images, and key learning is carried out on key parts, so that subsequent key feature extraction and preliminary judgment are facilitated. The problems of low artificial recognition rate, scattered image data, low data utilization rate and the like are solved. And efficient and rapid identification of engine nondestructive testing images is realized. The method can be used for engine nondestructive testing image recognition, and compared with an existing method, the recognition efficiency and accuracy are remarkably improved. 本发明提供一种基于机器学习与计算机视觉固体火箭发动机无损检测方法，通过将输入的图像分为多个局部图像，对重点部位进行重点学习，方便后续提取关键特征并做出初步判断。本发明解决人工识别率低、图像数据分散、数据利用率低等问题。实现发动机无损检测图像高效、快速识别。本发明可作为发动机无损检测图像识别使用，相比于现有方法，识别效率与准确率显著提高。