自然背景下计算机视觉技术在西红柿采摘中的应用 |
| |
引用本文: | 李思广.自然背景下计算机视觉技术在西红柿采摘中的应用[J].农机化研究,2017(7):201-203. |
| |
作者姓名: | 李思广 |
| |
作者单位: | 周口职业技术学院,河南周口,466000 |
| |
基金项目: | 河南省高等学校重点科研项目(15B520047) |
| |
摘 要: | 针对西红柿种植过程中,采摘费时费力的问题,将计算机视觉技术与图像处理和模式识别技术相结合,开发了在自然背景条件下的西红柿自动识别系统。运用双目立体视觉系统来识别红色西红柿,去除原始图像的噪声、转换图像的灰度、对新图像进行分割和目标特征提取,从背景中把西红柿分离出来;同时,针对果实间存在的重叠问题,采用圆形Hough变换算法,提取西红柿的圆心坐标和半径特征,通过获取T=Sqrt(s×l)图像,恢复被遮挡西红柿。利用该技术,采集了不同条件下的100张西红柿照片,其中99张照片被准确识别,识别率达到9 9%以上,能够较好地满足西红柿采摘工作的要求。
|
关 键 词: | 双目立体视觉 自动识别 定位 西红柿 采摘 |
Application of Computer Vision Technology in Tomato Picking under Natural Background |
| |
Abstract: | Computer vision technology is an important branch of computer technology .It has been widely used and devel-oped from the beginning of last century .According to the picking time-consuming problem during tomato planting , the combination of computer vision technology and image processing and pattern recognition technology , developed under nat-ural background condition of tomato automatic identification and picking method .Using binocular stereo vision system to i-dentify the red tomatoes , removal of the original image noise ,image conversion of gray ,the new image of segmentation and target feature extraction , from the background to isolate the tomatoes;and according to the fruit of overlapping problem , using circular Hough transform algorithm , extraction tomatoes of the center coordinates and radius of the characteristics , through T=Sqrt (s x L) image acquisition, the restoration of occlusion tomatoes .Using this technique,100 tomato pic-tures were collected under different conditions , the 99 pictures were identified accurately and the recognition rate reached 99%, which could meet the requirements of the tomato picking .The development and application of this technology can improve the picking efficiency and reduce the production cost . |
| |
Keywords: | binocular stereo vision automatic recognition orientation tomato picking |
本文献已被 CNKI 万方数据 等数据库收录! |
|