Improved genetic algorithm combined with improved Otsu algorithm for field crop segmentation

doi:10.14088/j.cnki.issn0439-8114.2019.15.028

Abstract

Abstract: For some field images, it is difficult to determine the optimal threshold problem of image segmentation due to its complicated background and uneven illumination. This paper proposes an image segmentation method based on improved Otsu algorithm optimization and improved genetic algorithm. Firstly, the acquired images are pre-processed. Based on the preprocessed images, the genetic control parameters can be automatically adjusted by improving the three methods of selection, crossover and variation in the genetic algorithm and optimizing the individual fitness function based on Otsu, so as to ensure the diversity of species and accelerate its convergence speed. The optimal threshold is provided for the Otsu image segmentation, and finally the image is filled by image morphology. In the result of the discussion, the algorithm results are compared with the Genetic Algorithm Based on the Otsu Algorithm and the Image Segmentation Based on Genetic Algorithm and KSW Entropy. It is found that the threshold range obtained by the algorithm is stable, which makes the segmented image accurate and clear. It is helpful to calculate the number of crops or the coverage of plants in the later stage.

Key words: global threshold segmentation, genetic algorithm, Otsu algorithm

CLC Number:

TP751

ZHAO Ming-xia, LYU Zhi, HAO Ya-jie, SHI Wei-jie, LI Fu-zhong. Improved genetic algorithm combined with improved Otsu algorithm for field crop segmentation[J]. HUBEI AGRICULTURAL SCIENCES, 2019, 58(15): 119-123.

References

[1] 谭优, 王泽勇.图像阈值分割算法实用技术研究与比较[J].微计算机信息, 2007, 23(24):298-299, 233.
[2] SHARMA S, SHAH D.A practical animal detection and collision avoidance system using computer vision technique[J].IEEE Access, 2016(99):1.
[3] HORNG M H.Multilevel thresholding selection based on the artificial bee colony algorithm for image segmentation[J].Expert Syst Appl, 2011, 38(11):13785-13791.
[4] ABUTALEB AS.Automatic thresholding of gray-level pictures using two-dimensional entropy[J].Computer vision graphics and image processing, 1989, 47(1):22-32.
[5] CHANG D, ZHAO Y, LIU L, et al.A dynamic niching clustering algorithm based on individual-connectedness and its application to color image segmentation[J].Pattern recognition, 2016, 60: 334-347.
[6] 刘桂红, 赵亮, 孙劲光, 等.一种改进粒子群优化算法的Otsu图像阈值分割方法[J].计算机科学, 2016, 43(3):309-312.
[7] KIRKPATRICK S.Optimization by simulated annealing: quantitative studies[J].J Stat Phys, 1984, 34(5-6):975-986.
[8] SOURAV DE, BHATTACHARYYA S, PARAMARTHA D.Automatic magnetic resonance image segmentation by fuzzy intercluster hostility index based genetic algorithm:An application[J].Applied soft computing, 2016, 47:669-683.
[9] JUANG C F.A hybrid of genetic algorithm and particle swarm optimization for recurrent network design IEEE Trans[J].Syst Man Cybern B:Cybern, 2004, 34(2):997-1006.
[10] 姚立平, 潘中良.使用遗传算法和KSW熵法相结合的CT图像分割[J].电视技术, 2018, 42(11):1-6.
[11] KITTLER J, ILLINGWORTH J.Minimum error thresholding[J]. Pattern recognition, 1986, 19(1):41-47.
[12] OTSU N.A threshold selection method from gray-level histogram IEEE Trans[J].Syst Man Cybern, 1979(9):62-66.
[13] 张东生. 基于阈值的图像分割算法研究[D].黑龙江大庆:东北石油大学, 2011.
[14] HOLLAND J H.Adaptation in natural and artificial systems:An introductory analysis with applications to biology, control, and artificial intelligence[M].East lansing:University michigan press, 1975.
[15] AZAMATHULLA H M, WU F C, AB GHANI A, et al.Comparison between genetic algorithm and linear programming approach for real time operation[J].J Hydro-environ Res, 2008, 2(3):172-181.