基于图像处理的白酒酒花轮廓检测

依据看花摘酒的传统经验,采用机器视觉代替人眼,通过CCD获取摘酒酒花的视频图像,并截取不同酒度酒花图像进行直方图均衡化、图像腐蚀等图像预处理,消除了高光噪声的影响,然后采用不同边缘检测算法对酒花轮廓进行了对比研究,采用OTSU算法与Canny边缘检测算法相结合的方法,较好地实现酒花与背景的分割,提取清晰的酒花边缘轮廓,通过对大清花与小清花图像的模式识别,为摘酒自动化提供了有效分级依据。该智能化的分级摘酒方法,能够提高分级摘酒工艺的稳定性和准确性,易于实现分级摘酒工序的智能自动化。

Liquor hop contour detection based on image processing

Based on the traditional experience of liquor-receiving according to liquor hop, Computer vision is used instead of human eyes, the video images of liquor-receiving is captured by CCD, and the images of hops with different liquor degrees are preprocessed by histogram equalization and image corrosion to eliminate the influence of high light noise. Then hops contour is compared with different edge detection algorithms. Besides, the combination of OTSU algorithm and Canny edge detection algorithm can better realize the segmentation of hops and background, and clear edge contour is extracted. Through the pattern recognition of Daqing flower and Xiaoqing flower images, an effective grading basis for liquor-receiving automation is provided. This intelligent grading method can improve the stability and accuracy of the graded liquor-receiving process, and it is easy to realize the intelligent automation of the graded liquor-receiving process.