This study regards automatic interpretation of the pointer meter and improving the identification precision of the verification system as the goal.In this paper a new method is put forward,which makes full use of the water-meter image structure information to identify water-meter readings.The advantages of such a recognition system are as follows.First,the system applicable scope is so broad that any red basal pointer can realize the extraction of pointer.Extraction process,which uses the pointer LRCD model about color difference to process images,improves the discriminating accuracy about pointer-image.Second,for the determination of the direction of the pointer,this study abandons the traditional methods of model matching and adopts the way of seeking the centroid of needlepoint`s pixels.Finally,this paper improves angle method to realize the improvement of the decipherable accuracy,owing to the fact that it regards the indication error in different flow point of the system as the main verify indicators.The experimental results show that,compared with other traditional methods,the new one which uses the image processing and the modified angle identification recognition method has good accuracy and reliability.%为实现指针式水表自动判读和提高识别精度,结合水表表盘结构信息来识别表盘读数的图像.水表指针提取过程中采用色差模型对图像进行处理;计算针尖像素质心均值实现对指针方向的确定;特别对指针判读时采用一种改进的角度法实现判读准确性的提高.最后,以检定水表的不同流量点的示值误差作为衡量系统准确性的主要验证指标,通过与其他传统方法相比较,结果表明,利用图像处理和改进的角度判读方法具有很好的准确性和可靠性.
展开▼
