| 李靖宇， 沈丹峰， 王玉， 李耀杰.基于参数可调均值迁移滤波实现筒子纱的识别检测[J].轻工机械,2021,39(3):63-68 |
| 基于参数可调均值迁移滤波实现筒子纱的识别检测 |
| Recognition and Detection of Cheese Based on Edge Preserving Filtering Algorithm |
| |
| DOI：10.3969/j.issn.1005 2895.2021.03.011 |
| 中文关键词: 机器视觉 参数可调均值迁移滤波算法 锐化处理 边缘检测 几何矩 横纵比过滤 |
| 英文关键词:machine vision parameter adjustable mean shift filtering algorithm sharpening treatment edge detection geometric moment transverse longitudinal ratio filtration |
| 基金项目:陕西省科技厅重点研发计划（2018TSCXL GY 03 02）。 |
|
| 摘要点击次数: 683 |
| 全文下载次数: 836 |
| 中文摘要: |
| 为了提高机器视觉对纱筒图像识别检测的准确率，课题组提出了一种参数可调均值迁移滤波算法。在图像采样过程中，先对源图像进行锐化处理，再对图像空间窗口与颜色窗口同时进行卷积操作，筒子纱的边缘轮廓得到完整保留，且通过滤波处理去除图像噪声；采用Canny算子边缘检测，完成对筒子纱的轮廓提取;几何矩计算轮廓质心和横纵比过滤判别筒子纱的放置状态。实验结果表明：该方法在图像处理过程中准确、可靠，能够满足图像识别检测的要求。 |
| 英文摘要: |
| In order to improve the accuracy of machine vision on screen image recognition, the research group proposed a parameter adjustable mean shift filtering algorithm. In the process of image sampling, first the source image was sharpened, then the image space window and color window were convolved simultaneously. The edge contour of cheese was completely preserved, and the image noise was removed by filtering. Canny operator edge detection was used to complete the contour extraction of cheese; geometric moment calculation of contour centroid and transverse/longitudinal ratio filtering were used to judge the placement state of cheese. The experimental results show that this method is accurate and reliable in image processing, and can meet the requirements of image recognition and detection. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
