Actuator gear defect detection method based on image processing
By acquiring the gradient values and grayscale gradient continuity of the gear surface image, calculating the fracture intensity of the curve texture and adaptively updating the grayscale values, the problem of missed detection and false detection in gear defect detection by the traditional mean filtering algorithm is solved, and high-precision defect detection is achieved.
CN122415459APending Publication Date: 2026-07-17HAO YONG AUTOMOTIVE CONTROLS LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HAO YONG AUTOMOTIVE CONTROLS LTD
- Filing Date
- 2026-03-31
- Publication Date
- 2026-07-17
Smart Images
Figure CN122415459A_ABST
Abstract
本发明涉及图像处理技术领域，更具体地，本发明涉及基于图像处理的作动器齿轮缺陷检测方法，该方法包括，采集齿轮表面图像并获取每个像素点的水平和垂直方向梯度值；随后计算各像素点沿梯度方向的灰度渐变连续性，并结合单位正交法向量的分量推导出曲线纹理断裂强度，以此区分油膜与裂纹缺陷；利用所述曲线纹理断裂强度构建动态平滑保留调节因子，并基于该因子对原始灰度值与局部窗口均值进行自适应加权融合，生成更新灰度值。本发明实现了在强力平滑油膜干扰的同时精确保留微小缺陷边缘，显著提升了作动器齿轮缺陷检测的准确性与可靠性。
Need to check novelty before this filing date? Find Prior Art