Abstract
A method based on image processing, principal component analysis (PCA) and machine learning was proposed to identify pumpkin powdery mildew and improve the identification accuracy of pumpkin powdery mildew. The leaf-disease image was processed by the color feature compositing and detection method,in order to segment the lesion more accurately. Then 20 feature values are extracted, including the texture features, color features and morphological features of the segmented lesions, and the 20-dimensional original feature parameters were simplified to 3-dimensional feature parameters via the PCA method. Finally, three different SVM kernel functions are used to construct the classification model to compare the original feature parameters and principal component feature parameters separately. Experiment results show that the effect of illumination angle and intensity can be effectively eliminated after the super red feature calculation, and that the SVM model based on PCA and polynomial kernel function has a better recognition effect on pumpkin powdery mildew. The recognition rate is 97.3%, which is 3.15% higher than that of the traditional original characteristic parameters. Based on this finding, it can be concluded that this machine learning model can achieve accurate lesions identification.
Similar content being viewed by others
References
Bakhshipour A, Jafari A (2018) Evaluation of support vector machine and artificial neural networks in weed detection using shape features. Comput Electron Agric 145:153–160
Ding W, Taylor G (2016) Automatic moth detection from trap images for pest management. Comput Electron Agric 123:17–28
Durbha SS, King RL, Younan NH (2017) Support vector machines regression for retrieval of leaf area index from multiangle imaging spectroradiometer. Remote Sens Environ 107:348–361
Han DY, Huang XY, Fu H (2012) Measurement of plant leaf area based on image segmentation of color channel similarity. Trans Chin Soc Agric Eng 28(6):179–183
Han L, Haleem MS, Taylor M (2015) A novel computer vision-based approach to automatic detection and severity assessment of crop diseases. IEEE Science and Information Conference, 638–644
Hernández-Hernández JL, García-Mateos G, González-Esquiva JM, Escarabajal-Henarejos D, Ruiz-Canales A, Molina-Martínez JM (2016) Optimal color space selection method for plant/soil segmentation in agriculture. Comput Electron Agric 122:124–132
Jhuria M, Kumar A, Borse R (2013) Image processing for smart farming: Detection of disease and fruit grading. Image Information Processing (ICIIP). IEEE International Conference on Image Information Processing, 521–526
Jhuria M, Kumar A, Borse R (2013) Image processing for smart farming: Detection of disease and fruit grading. IEEE Image Information Processing (ICIIP).International Conference on Image Information Processing, 521–526
Kruse OMO, Prats-Montalbán JM, Indahl UG, Kvaal K, Ferrer A, Futsaether CM (2014) Pixel classification methods for identifying and quantifying leaf surface injury from digital images. Comput Electron Agric 108:55–165
Liu W, Wang ZD, Liu XH, Zeng NY, Liu YR, Alsaadi FE (2017) A survey of deep neural network architectures and their applications. Neurocomputing 234:11–26
Liu Y, Wang R, Zhang J (2019) Accurate identification method of powdery mildew in colored pepper based on convolutional neural network. Inf Comput (Theoretical Ed) (07):40–43
Ma JC, Du KM, Zhang LX, Zheng FX, Chu JX, Sun ZF (2017) A segmentation method for greenhouse vegetable foliar disease spots images using color information and region growing. Comput Electron Agric 142:110–117
Meyer GE, Hindman TW, Laksmi K (1999) Machine vision detection parameters for plant species identification. Precision Agric Biol Qual Int Soc Opt Photon 3543:327–335
Minervini M, Abdelsamea MM, Tsaftaris SA (2014) Image-based plant phenotyping with incremental learning and active contours. Ecol Inform 23:35–48
Phadikar S, Sil J (2016) Rice disease identification using pattern recognition techniques. IEEE International Conference on Computer and Information Technology, 420–423
Phadikar S, Sil J (2018) Rice disease identification using pattern recognition techniques. IEEE International Conference on Computer and Information Technology, 420–423
Pujari JD, Yakkundimath R, Byadgi AS (2015) Image processing based detection of fungal diseases in plants. Procedia Comput Sci 46:1802–1808
Ren SG, Lu HF, Yuan PS (2016) Segmentation algorithm of cucumber leaf disease image based on saliency detection. Trans Chin Soc Agric Mach 47(9):11–16
Sankaran S, Mishra A, Ehsani R, Davis C (2010) A review of advanced techniques for detecting plant diseases. Comput Electron Agric 72(1):1–13
Singh V, Varsha, Misra AK (2015) Detection of unhealthy region of plant leaves using image processing and genetic algorithm. IEEE Computer Engineering and Applications, 1028–1032
Tanzeel U, Rehman M, Sultan M, Young K, Chang JJ, Jaemyung S (2019) Current and future applications of statistical machine learning algorithms for agricultural machine vision systems. Comput Electron Agric 156:585–605
Uzunbas MG, Chen C, Metaxas D (2016) An efficient conditional random field approach for automatic and interactive neuron segmentation. Med Image Anal 27:31–44
Wang XY, Zhang XJ, Yang HY (2012) A pixel-based color image segmentation using support vector machine and fuzzy -means. Neural Netw 33(9):148–159
Wang XY, Wen HJ, Li XX (2016) Research progress analysis of mainly agricultural diseases detection and early warning technologies. Trans Chin Soc Agric Mach 47(9):266–277
Wei WH, Li Yl, Xue TM, Li YT, Sun P, Yang B, Yin Z, Mei CT (2019) Tool wear during high-speed milling of wood-plastic composites. Bioresources 14(4):8678–8688
Wu DY, Ye N, Shen LR, Zhang QQ, Lai ZW (2019) Clustering method for automatic timber defects detection based on the color moment. J Jiangnan Univ (Nat Sci Ed) 8(5):520–524
Xu XM, Jiang YP, Heow PL (2017) Multi-objective optimal design of sandwich panels usingagenetic algorithm. Eng Optim 49(10):1665–1684
Xue Z, Ming D, Song W, Wan BK, Jin SJ (2010) Infrared gait recognition based on wavelet transform and support vector machine. Pattern Recogn 43(8):2904–2910
Ye Q, Jiang X, Li X et al (2017) Comparison on inversion model of soil organic matter content based on hyperspectral data. Trans Chin Soc Agric Mach 48(3):164–172
Zhang DY, Xie XH, Ling D (2016) Comparative analysis of identification methods for wheat leaf diseases. Int Agric Eng J 25(3):197–203
Zhang S, Wu X, You Z, Zhang LQ (2017) Leaf image based cucumber disease recognition using sparse representation classification. Comput Electron Agric 134:135–141
Zhang SW, Wang HX, Huang WZ, You ZH (2018) Plant diseased leaf segmentation and recognition by fusion of superpixel K-means PHOG. Optik 157:866–872
Funding
The research is funded by Jiangsu Provincial Natural and Science Founding (BK20170727), National Natural Science Foundation of China (61701242) and Special Plan for Innovation and Entrepreneurship Training for College Students of Nanjing Agricultural University(S20190036).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that there are no competing interests regarding the publication of this paper.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Lin, H., Sheng, H., Sun, G. et al. Identification of pumpkin powdery mildew based on image processing PCA and machine learning. Multimed Tools Appl 80, 21085–21099 (2021). https://doi.org/10.1007/s11042-020-10419-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-020-10419-1