利用色彩局部二元圖形於植物疾病檢測

Detecting Plant Diseases Using Color Local Binary Patterns

張峻瑋(Chun-Wei Chang)

指導教授 ： 黃惠藩

亞洲大學/資訊學院/資訊工程學系碩士班/碩士(2011年)

摘要

植物病害在全球農業造成主要農作物的經濟損失，本論文提供自動以及低成本的檢測方法。目前大部分的方法成本高，例如高光譜成像以及紅外線光譜和螢光光譜儀還有核磁共振，這些都需要高階硬體的配備，大大增加成本。本論文提出一種利用色彩局部二元圖形(LBP)來檢測植物疾病的方法。局部二元圖形提供高度可區別的紋理資訊，並且不會受光照亮度之影響，本方法運用像素之色彩加上從像素四周抽取出之色彩局部二元圖形特性來分割出感染疾病的區域。實驗結果顯示，運用色彩局部二元圖形紋理特徵並結合支持向量機(SVM)分類器能有效的在植物色彩圖片中切割出疾病區域。此實驗結果使用像素色彩加紋路特徵比單獨做色彩與單獨做紋路特徵有效。

關鍵字

疾病檢測 ； 植物 ； 色彩紋理 ； 局部二元圖形

並列摘要

This study presents a method for detecting plant diseases using color Local Binary Patterns (LBP). LBP provides highly discriminative texture information and is invariant to any monotonic changes in gray level. The proposed method uses pixel color and the LBP features extracted from a region surrounding a pixel to segment the diseased regions. Experimental results show that color LBP texture features combining with Support Vector Machine (SVM) classifier are effective for segmenting diseased regions in plant color images.

並列關鍵字

Disease Detection ； Plant ； Color Texture ； Local Binary Pattern

參考文獻

[7] C. Campbell and L. Lucas B. Lucas, Introduction to Plant Diseases: Identification and Management, Van Nostrand Reinhold, U.S., 1992.

[1] A. Mishra, R. Ehsani and C. Davis S. Sankaran, "A review of advanced techniques for detecting plant disearses," Computers and Electronics in Agriculture, vol. 72, no. 1, pp. 1-13, 2010.

[2] A. Camargo and J. Smith, "An image-processing based algorithm to automatically identify plan disease visual symptoms," Biosystems Engineering, vol. 102, pp. 9-21, 2009.

[4] M. Gasparoto and L. Marcassa L. Belasque, "Detection of mechanical and disease stresses in citrus plants by fluorescence spectroscopy," Applied Optics, vol. 47, no. 11, pp. 1922-1926, 2008.

5] H. Shafri and N. Hamdan, "Hyperspectral imagery for mapping disease infection in oil palm plantation using vegetation indices and red edge techniques," American Journal of Applied Sciences, vol. 6, no. 6, pp. 1031-1035, 2009.