AUTHOR=Lu Yang , Du Jiaojiao , Liu Pengfei , Zhang Yong , Hao Zhiqiang 

TITLE=Image Classification and Recognition of Rice Diseases: A Hybrid DBN and Particle Swarm Optimization Algorithm

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2022.855667

DOI=10.3389/fbioe.2022.855667

ISSN=2296-4185

ABSTRACT=Rice blast, rice sheath blight and rice brown spot have become the most popular diseases in
the cold areas of northern China. In order to further improve the accuracy and efficiency of rice
diseases diagnosis, a framework for automatic classification and recognition of rice diseases is
proposed in this paper. Firstly, we construct a training and testing data set including 1500 images
of rice blast, 1500 images of rice sheath blight, 1500 images of rice brown spot and 1100 healthy
 images were collected from the rice experimental field. Secondly, the deep belief network (DBN)
model is designed including 15 hidden restricted Boltzmann machines layers and support vector
 machine (SVM) optimized with switching particle swarm (SPSO). It is noted that the developed
 DBN and SPSO-SVM can simultaneously learn three proposed features including color, texture,
shape to recognize the disease type from the region of interest that obtained by pre-processing
the disease images. The proposed model leads to hit rate 91.37%, accuracy 94.03% and false
measurement rate 8.63% respectively with the 10-fold cross-validation strategy. The value of area
under the receiver operating characteristic curve (AUC) is 0.97 whose accuracy is much higher
than that of conventional machine learning model. The simulation results show that DBN and
SPSO-SVM models can effectively extract the image features of rice diseases during recognition,
and have good anti-interference and robustness.