Hyperspectral system for early detection of rottenness caused by Penicillium digitatum in mandarins

被引:130
作者
Gomez-Sanchis, J. [1 ]
Gomez-Chova, L. [2 ]
Aleixos, N. [3 ]
Camps-Valls, G. [2 ]
Montesinos-Herrero, C. [1 ]
Molto, E. [1 ]
Blasco, J. [1 ]
机构
[1] Inst Valenciano Invest Agr, Ctr Agroingn, Valencia 46113, Spain
[2] Univ Valencia, Dept Elect Engn, GPDS, Digital Signal Proc Grp, E-46100 Valencia, Spain
[3] Univ Politecn Valencia, Dept Graph Engn, DIG ETSII, Valencia 46022, Spain
关键词
fruit inspection; mandarins; feature selection; hyperspectral imaging; machine vision; image analysis; CART; LDA;
D O I
10.1016/j.jfoodeng.2008.04.009
中图分类号
TQ [化学工业];
学科分类号
0817 ;
摘要
Nowadays, the detection of fruit infected with Penicillium sp. fungi on packing lines is carried out manually under ultraviolet illumination. Ultraviolet sources induce visible fluorescence of essential oils present in the skin of citrus and which are released by the action of fungi, thus increasing the contrast between sound and rotten skin. This work analyses a set of techniques aimed at detecting rotten citrus without the use of UV lighting. The techniques used include hyperspectral image acquisition, preprocessing and calibration, feature selection and segmentation using linear and non-linear methods for classification of fruits. Different methods such as correlation analysis, mutual information, stepwise, and genetic algorithms based on linear discriminant analysis (LDA) are studied to select the most relevant bands. image segmentation relies on the combination of efficient band selection techniques and also on pixel classification methods such as classification and regression trees (CART) and LDA. The results were obtained using a large dataset of images of mandarins cv. "Clemenules" by applying the CART method. The hyperspectral computer vision system proposed here is capable of detecting damage caused by Penicillium digitatum in mandarins using a reduced set of optimally selected bands. (c) 2008 Elsevier Ltd. All rights reserved.
引用
收藏
页码:80 / 86
页数:7
相关论文
共 31 条
[1]   Multispectral inspection of citrus in real-time using machine vision and digital signal processors [J].
Aleixos, N ;
Blasco, J ;
Navarrón, F ;
Moltó, E .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2002, 33 (02) :121-137
[2]  
[Anonymous], 1989, CITRUS IND, DOI DOI 10.1016/j.postharvbio.2008.11.008
[3]  
BAJKSY P, 1997, PREDICTION ACCURACY
[4]   USING MUTUAL INFORMATION FOR SELECTING FEATURES IN SUPERVISED NEURAL-NET LEARNING [J].
BATTITI, R .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (04) :537-550
[5]   Citrus sorting by identification of the most common defects using multispectral computer vision [J].
Blasco, J. ;
Aleixos, N. ;
Gomez, J. ;
Molto, E. .
JOURNAL OF FOOD ENGINEERING, 2007, 83 (03) :384-393
[6]   Computer vision detection of peel defects in citrus by means of a region oriented segmentation algorithm [J].
Blasco, J. ;
Aleixos, N. ;
Molto, E. .
JOURNAL OF FOOD ENGINEERING, 2007, 81 (03) :535-543
[7]   Selection of relevant features and examples in machine learning [J].
Blum, AL ;
Langley, P .
ARTIFICIAL INTELLIGENCE, 1997, 97 (1-2) :245-271
[8]  
Chang C.-I., 2003, Hyperspectral Imaging: Techniques for Spectral Detection and Classification, V1
[9]  
Duda R. O., 1973, Pattern Classification
[10]  
Evans MD, 1998, T ASAE, V41, P1845, DOI 10.13031/2013.17324