【摘要】 针对食品包装行业中喷墨打印技术形成的点阵字符不连续及食品包装背景复杂等问题,提出一种基于改进的连通域分割与BP神经网络相结合的字符识别方法。利用数学形态学与连通域分析定位点阵字符区域,对字符区域进行分割,利用BP神经网络对字符进行识别。实验结果表明,该方法在字符分割和识别上的效果均有较大的提升,其字符识别准确率达到98.03%。更多还原

【Abstract】 Aiming at the problems of dot matrix character discontinuity formed by the inkjet technology and the complex background of food packaging, a character recognition method based on improved connected domain segmentation and BP neural network is proposed. The mathematical morphology and connected domain analysis were used to locate the lattice character region, and the character regions were divided. The BP neural network was used to identify the characters. The experimental results show that this method has a great improvement in character segmentation and recognition effect, and its recognition accuracy achieves 98.03%.更多还原