PDF417 Angle Detection under Complex Background Based On Morphology and Genetic Algorithms Work
Guocheng Wang, LI Jian-hua, Chen Yi, Meng Zhang, Mao Yincan, Li Ping · Procedia Engineering · 2012
The automatic identification of PDF417 bar code is very sensitive to skew angle. However, the common skew angle detection methods have shortcomings such as computationally expensive or high complexity. In this paper, morphology is used to extract PDF417 from complex background, with PDF417's character that the start symbol, the stop symbol and the edge of the module presents straight lines. Firstly, utilize opening operation to detect the straight lines. Then, Combined with genetic algorithm, these straight lines are extracted, and the angle is obtained. Here, classical genetic algorithm is improved combined with simulated annealing algorithm to make algorithm converge faster and not so easily to fall into local optimal solution. The trial picture's angle is 110.13°, the time spending on Process keeps to about only 0.40 s for different images, and the precision is basically around 0.1°. Experimental results show that the algorithm has the advantage of less computation and high precision.