Do we really need deep CNN for plant diseases identification?

Li, Yang; Nie, Jing; Chao, Xuewei

doi:10.1016/j.compag.2020.105803

Cited by 140 publications

(49 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Corresponding color cameras are configured in the original reconstruction system. In order to ensure that the collected Complexity initial information is suitable for visual communication technology, the depth camera is used on the original basis to complete the acquisition of depth data [27][28][29][30]. e data I/O submodule, parameter setting submodule, volume rendering submodule, surface rendering submodule, and axial slice display submodule together constitute the 3D visualization module, and its structure is shown in Figure 1.…”

Section: Level Distribution Of Visualmentioning

confidence: 99%

Virtual Reconstruction Method of Regional 3D Image Based on Visual Transmission Effect

Yin

Yang

2021

Complexity

View full text Add to dashboard Cite

In order to solve the problems of poor user experience and low human-computer interaction efficiency, this paper designs a 3D image virtual reconstruction system based on visual communication effects. First, the functional framework diagram and hardware structure diagram of the image 3D reconstruction system are given. Then, combined with the basic theory of visual communication design, the characteristics of different elements in the three-dimensional image system and reasonable visual communication forms are analyzed, and design principles are proposed to improve user experience and communication efficiency. After the input image is preprocessed by median filtering, a three-dimensional reconstruction algorithm based on the image sequence is used to perform a three-dimensional reconstruction of the preprocessed image. The performance of the designed system was tested in a comparison form. We optimize the original hardware structure, expand the clock module, and use the chip to improve the data processing efficiency; in the two-dimensional image; we read the main information, through data conversion, display it in three-dimensional form, select the feature area, extract the image feature, calculate the key physical coordinate points, complete the main code compilation, use visual communication technology to feed back the display visual elements to the 3D image, and complete the design of the 3D image virtual reconstruction system. The test results showed that the application of visual communication technology to the virtual reconstruction of 3D images can effectively remove noise and make the edge area of the image clear, which can meet the needs of users compared with the reconstruction results of the original system. Visual C++ and 3DMAX are used as the system design platform, and three-dimensional image visualization and roaming are realized through OpenGL. Experimental results show that the designed system has better reconstruction accuracy and user satisfaction.

show abstract

Section: Level Distribution Of Visualmentioning

confidence: 99%

Virtual Reconstruction Method of Regional 3D Image Based on Visual Transmission Effect

Yin

Yang

2021

Complexity

View full text Add to dashboard Cite

show abstract

“…In recent years, some deep learning algorithms such as convolutional neural networks (CNN) and machine learning algorithms such as support vector machine regression (SVR), partial least squares regression (PLSR), neural network, and random forest (RF) have been applied to agricultural condition monitoring, plant disease and insect monitoring, wheat ear identification and other aspects, and have shown good results. For example, Li et al [ 16 , 17 ] used CNN to carry out identification and monitoring of plant diseases and insect pests. Xu et al [ 18 ] used CNN to achieve accurate segmentation and recognition of the number of wheat ears.…”

Section: Introductionmentioning

confidence: 99%

Leaf area index estimation model for UAV image hyperspectral data based on wavelength variable selection and machine learning methods

et al. 2021

View full text Add to dashboard Cite

Background To accurately estimate winter wheat leaf area index (LAI) using unmanned aerial vehicle (UAV) hyperspectral imagery is crucial for crop growth monitoring, fertilization management, and development of precision agriculture. Methods The UAV hyperspectral imaging data, Analytical Spectral Devices (ASD) data, and LAI were simultaneously obtained at main growth stages (jointing stage, booting stage, and filling stage) of various winter wheat varieties under various nitrogen fertilizer treatments. The characteristic bands related to LAI were extracted from UAV hyperspectral data with different algorithms including first derivative (FD), successive projections algorithm (SPA), competitive adaptive reweighed sampling (CARS), and competitive adaptive reweighed sampling combined with successive projections algorithm (CARS_SPA). Furthermore, three modeling machine learning methods including partial least squares regression (PLSR), support vector machine regression (SVR), and extreme gradient boosting (Xgboost) were used to build LAI estimation models. Results The results show that the correlation coefficient between UAV and ASD hyperspectral data is greater than 0.99, indicating the UAV data can be used for estimation of wheat growth information. The LAI bands selected by using different algorithms were slightly different among the 15 models built in this study. The Xgboost model using nine consecutive characteristic bands selected by CARS_SPA algorithm as input was proved to have the best performance. This model yielded identical results of coefficient of determination (0.89) for both calibration set and validation set, indicating a high accuracy of this model. Conclusions The Xgboost modeling method in combine with CARS_SPA algorithm can reduce input variables and improve the efficiency of model operation. The results provide reference and technical support for nondestructive and rapid estimation of winter wheat LAI by using UAV.

show abstract

“…The other way is to solve the classification problem with few data, also called few-shot learning, which is more suitable for practical applications. For example, some other works focused on model compression by pruning [ 23 ], shallow model [ 24 ], and lightweight network [ 25 ].…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised few-shot learning approach for plant diseases recognition

Chao

2021

Plant Methods

Self Cite

View full text Add to dashboard Cite

Background Learning from a few samples to automatically recognize the plant leaf diseases is an attractive and promising study to protect the agricultural yield and quality. The existing few-shot classification studies in agriculture are mainly based on supervised learning schemes, ignoring unlabeled data's helpful information. Methods In this paper, we proposed a semi-supervised few-shot learning approach to solve the plant leaf diseases recognition. Specifically, the public PlantVillage dataset is used and split into the source domain and target domain. Extensive comparison experiments considering the domain split and few-shot parameters (N-way, k-shot) were carried out to validate the correctness and generalization of proposed semi-supervised few-shot methods. In terms of selecting pseudo-labeled samples in the semi-supervised process, we adopted the confidence interval to determine the number of unlabeled samples for pseudo-labelling adaptively. Results The average improvement by the single semi-supervised method is 2.8%, and that by the iterative semi-supervised method is 4.6%. Conclusions The proposed methods can outperform other related works with fewer labeled training data.

show abstract

Do we really need deep CNN for plant diseases identification?

Cited by 140 publications

References 19 publications

Virtual Reconstruction Method of Regional 3D Image Based on Visual Transmission Effect

Virtual Reconstruction Method of Regional 3D Image Based on Visual Transmission Effect

Leaf area index estimation model for UAV image hyperspectral data based on wavelength variable selection and machine learning methods

Semi-supervised few-shot learning approach for plant diseases recognition

Contact Info

Product

Resources

About