一种改进的卷积神经网络的数显仪表识别方法
An Improved Digital Display Instrument Recognition Method Based on Convolutional Neural Network
摘要: 为了提高数显仪表的识别率，设计了一种传统图像处理方法和深度学习技术相结合的算法，即一种基于改进的卷积神经网络的数显仪表识别算法。首先通过传统图像处理技术对图像进行图像预处理、字符分割等操作，然后由基于注意机制的卷积神经网络算法对字符进行识别。实验结果表明，该方法不仅有效提高了字符的准确率，字符识别率高达98.5%，还提高了网络的收敛速度。该方法基本可以满足各种数显仪表的识别，能够满足实际应用的需求。
Abstract: In order to improve the recognition rate of digital display instruments, an algorithm combining traditional image processing method and deep learning technology is designed, that is, an algorithm for digital display instrument recognition based on an improved convolutional neural network. Firstly, the traditional image processing technology is used to perform image preprocessing, character segmentation and other operations, and then the characters are recognized by the convolutional neural network algorithm based on the attention mechanism. The experimental results show that this method not only improves the character accuracy effectively, the character recognition rate is up to 98.5%, but also improves the convergence speed of the network. This method can basically meet the recognition of various digital display instruments and meet the requirements of practical application.
文章引用：李记花, 李鹤喜, 李威龙. 一种改进的卷积神经网络的数显仪表识别方法[J]. 计算机科学与应用, 2021, 11(2): 257-265. https://doi.org/10.12677/CSA.2021.112026

参考文献

[1] 王艳, 谢广苏, 沈晓宇. 一种基于MSER和SWT的新型车牌检测识别方法研究[J]. 计量学报, 2019, 40(1): 82-90.
[2] 张海波, 段会川, 张曙光, 等. 一种数字仪表显示值快速识别方法[J]. 计算机工程与应用, 2005, 41(4): 223-226.
[3] 王臣, 严军, 朱静, 等. 基于图像的数字仪表字轮定位分割算法的研究[J]. 电子测量技术, 2017, 40(9): 132-135.
[4] 申小阳, 唐轶峻, 姜柏军, 等. 仪表的数字字符识别系统[J]. 仪表技术与传感器, 2005(6): 55-57.
[5] 陆靖滨, 许丽. 基于自适应特征提取的数显仪表识别系统[J]. 现代电子技术, 2017, 40(24): 147-150.
[6] 蔡梦倩, 张蕾, 王炎, 莫娟. 基于全卷积网络的数字仪表字符识别方法[J]. 现代计算机(专业版), 2018(2): 38-43.
[7] Suzuki, S. (1985) Topological Structural Analysis of Digitized Binary Images by Border Follow-ing. Computer Vision, Graphics, and Image Processing, 30, 32-46. [Google Scholar] [CrossRef] 
[8] 丁双, 任国营, 张福民, 范乃胤. 改进的穿线法的卡尺图像识别[J]. 计量学报, 2019, 40(5): 765-769.
[9] 邓清男, 石晓龙. 变电站室内数显仪表的读数识别[J]. 工业仪表与自动化装置, 2018(2): 86-89.
[10] 卓磊, 周律, 杨丽红. 基于穿线法的计算器数字识别[J]. 软件工程, 2018, 21(12): 1-3.
[11] 陈刚, 胡子峰, 郑超. 基于特征检测的数字仪表数码快速识别算法[J]. 中国测试, 2019, 45(4): 146-150.
[12] 熊勋, 陈新度, 吴磊, 林旭华. 基于卷积记忆神经网络的数字表盘读数识别[J]. 组合机床与自动化加工技术, 2019(7): 72-75.
[13] Misra, D., Nalamada, T., Arasanipalai, A.U., et al. (2020) Rotate to Attend: Convo-lutional Triplet Attention Module.
[14] Chen, L., Zhang, H.W., Xiao, J., Nie, L.Q., Shao, J., Liu, W. and Chua, T.-S. (2017) Sca-cnn: Spatial and Channel-Wise Attention in Convolutional Networks for Image Captioning. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 21-26 July 2017, 5659-5667. [Google Scholar] [CrossRef] 
[15] Hu, J., Shen, L. and Sun, G. (2018) Squeeze-and-Excitation Net-works. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-22 June 2018, 7132-7141. [Google Scholar] [CrossRef] 
[16] Liu, J.-J., Hou, Q., Cheng, M.M., Wang, C.H. and Feng, J.S. (2020) Improving Convolutional Networks with Self-Cali- brated Convolutions. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, 13-19 June 2020, 10096-10105.
[17] Woo, S., Park, J., Lee, J.-Y. and Kweon, I.S. (2018) Cbam: Convolutional Block Attention Module. The European Conference on Computer Vision (ECCV), Munich, 8-14 September 2018, 3-19.