Abstract How to accurately identify the jujube tree diseases based on digital image processing is an unsolved problem. An algorithm proposed that was developed to identify the six varieties (jujube rust, jujube anthracnose, jujube white rot, jujube fruit rust disease, ascochyta spot of jujube, jujube witches broom) jujube trees diseases which are common. The algorithm was based on color, morphological features and texture features. Nine colors, eleven morphological and four texture features were used for analysis. Thirty pictures are used as the training data set and fifteen as the test data set in the neural network used to identify jujube trees diseases. When the model was tested on the test data set, the identification accuracies were 91.00%, 89.00%, 94.00%, 84.00%, 73.00%, 81.00% for jujube rust, jujube anthracnose, jujube white rot, jujube fruit rust disease, ascochyta spot of jujube, jujube witches broom, respectively.
[1]
F. S. Lai,et al.
Discrimination between wheat classes and varieties by image analysis
,
1986
.
[2]
Jiang Song.
Study on identification of rice varieties using computer vision
,
2004
.
[3]
Tony F. Chan,et al.
Image processing and analysis
,
2005
.
[4]
Digvir S. Jayas,et al.
CLASSIFICATION OF CEREAL GRAINS USING MACHINE VISION: IV. COMBINED MORPHOLOGY, COLOR, AND TEXTURE MODELS
,
2000
.
[5]
Samir Majumdar,et al.
Classification of cereal grains using machine vision
,
1997
.
[6]
Mehmed Kantardzic,et al.
Data Mining: Concepts, Models, Methods, and Algorithms
,
2002
.