摘要
针对传统遗传算法在食品二维码识别和追溯的应用中表现出识别准确性不高的问题,本文提出了一种基于算子和适应度函数优化遗传算法的食品二维码识别追溯模型,首先在遗传算法运行中依据种群的特点来动态调整交叉概率和变异概率的数值,然后采用一种将海明距离测度与适应度距离相结合的方法,将个体的目标适应度降低,最后将食品二维码识别追溯模型转化为图像识别模型,并采用改进的遗传算法对其进行二维码识别和追溯。仿真试验结果表明,本文提出的基于算子和适应度函数优化遗传算法的食品二维码识别追溯模型相比较标准遗传算法,具有更高的识别精度。
According to the low recognition accuracy of the traditional genetic algorithm in food QR codeidentification and tracing,a traceability model for food QR code identification is proposed based ongenetic algorithm with the optimization of operator and fitness function.First of all,in the geneticalgorithm running,dynamically adjust the value of crossover probability and mutation probability basedon the characteristics of the population.Then a method of combining hamming distance measure andfitness distance is used to reduce the individual fitness of the goals.Finally food QR code identificationmodel translated into image recognition model,and the improved genetic algorithm is for the QR codeidentifying and retrospect.Simulations show that compared with the standard genetic algorithm,theproposed model based on the improved algorithm has higher identification accuracy.
作者
安进
张丹
An Jin;Zhang Dan(Department of Information Engineering,Jiangsu Food & Pharmaceutical Science College,Huaian Jiangsu 223003,China)
出处
《科技通报》
北大核心
2017年第5期134-137,共4页
Bulletin of Science and Technology
基金
淮安市科技计划项目HAG20214014
关键词
食品追溯
二维码识别
算子优化
适应度函数
遗传算法
food tracing
QR code identification
operator optimization
fitness function
genetic algorithm