Identification of Navel Orange Diseases and Pests Based on the Fusion of DenseNet and Self-Attention Mechanism

Abstract

The prevention and control of navel orange pests and diseases is an important measure to ensure the yield of navel oranges. Aiming at the problems of slow speed, strong subjectivity, high requirements for professional knowledge required, and high identification costs in the identification methods of navel orange pests and diseases, this paper proposes a method based on DenseNet and attention. The power mechanism fusion (DCPSNET) identification method of navel orange diseases and pests improves the traditional deep dense network DenseNet model to realize accurate and efficient identification of navel orange diseases and pests. Due to the difficulty in collecting data of navel orange pests and diseases, this article uses image enhancement technology to expand. The experimental results show that, in the case of small samples, compared with the traditional model, the DCPSNET model can accurately identify different types of navel orange diseases and pests images and the accuracy of identifying six types of navel orange diseases and pests on the test set is as high as 96.90%. The method proposed in this paper has high recognition accuracy, realizes the intelligent recognition of navel orange diseases and pests, and also provides a way for high-precision recognition of small sample data sets.

Research on identification algorithm of crop pests and diseases based on improved DenseNet model

“…The disadvantage is that the sample size is small, which may lead to false positives or low average accuracy when targeting larger groups of more similar pests such as white flies and leaf moulds, for example, and is limited to tomato pests and diseases. Yin'e Zhang [3] et al proposed a navel orange pest identification method based on the fusion of Densenet and self-attention mechanism, which proposed a new DCPSNET network structure for navel orange pest species identification based on the DenseNet efficiency and attention mechanism. The advantages are the small size of the model and the high accuracy of the identification.…”

Image Recognition of Peach Pests Based on Improved Convnext

“…Compared with the semi-automatic intelligence of traditional machine learning methods, deep learning methods achieve end-to-end fully automatic intelligent recognition, which can automatically learn the basic features and deep semantic features of pest images from data, and are now widely used in the field of crop pest image recognition [4]. Zhang proposed a method based on the fusion of DenseNet and Self-attention mechanisms to achieve intelligent navel orange pest and disease The method is based on the fusion of DenseNet and Self-attention mechanisms [5]. Zhang proposed an improved extended residual network by introducing the residual convolution into the residual network without adding other parameters, which has the highest accuracy for the identification of stored grain pests compared to other methods [6].…”

Crop Pest Recognition in Real Agricultural Environment Using Convolutional Neural Networks by a Parallel Attention Mechanism

“…Guo et al (2020) designed a selfattention mechanism and incorporated it into the CNN structure, which achieved the optimal F1-scores of 93.21% for 11 types of crop diseases and pests. Zhang and Liu (2021) proposed a method based on DenseNet and an attention mechanism, and the model could identify 7 types of navel orange diseases and pests on the test set with 96.90% accuracy. The results in this study are compared with on other studies as summarized in Table 1.…”