摘要
水泵特性曲线是泵站设计、评价重要的基础资料。MATLAB软件具有强大图像处理功能,提出基于MATLAB平台编写选定曲线的自动追踪识别算法对水泵图像资料进行数值化。为保证曲线追踪识别的连续,可利用已识别曲线像素,寻求周围曲线像素并结合曲线斜率不可突变作为判断条件,甄别取舍寻求到的周围像素点。实践表明:采用该方法提高了水泵特性曲线提取的正确率,并且抗干扰能力强。
The pump characteristic curve is an important basic data for the design and evaluation of pumping stations.MATLAB is good at image processing.This paper proposes an automatic tracking and recognition algorithm based on the MATLAB platform to extract the curve data of the model pump.In order to ensure the continuity of curve recognition,the curve pixels can be selected combined searching for surrounding curve pixels with the non-abrupt change in the slope of the curve.Practice shows that this method improves the accuracy of model pump characteristic curve extraction and has strong anti-interference ability.
作者
丁聪
邸南思
DING Cong;DI Nansi(Shanghai Investigation, Design & Research Institute Co., Ltd, Shanghai 200335, China)
出处
《广东水利水电》
2021年第7期113-117,共5页
Guangdong Water Resources and Hydropower