1
|
Prince RH, Mamun AA, Peyal HI, Miraz S, Nahiduzzaman M, Khandakar A, Ayari MA. CSXAI: a lightweight 2D CNN-SVM model for detection and classification of various crop diseases with explainable AI visualization. FRONTIERS IN PLANT SCIENCE 2024; 15:1412988. [PMID: 39036360 PMCID: PMC11257924 DOI: 10.3389/fpls.2024.1412988] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 04/06/2024] [Accepted: 06/07/2024] [Indexed: 07/23/2024]
Abstract
Plant diseases significantly impact crop productivity and quality, posing a serious threat to global agriculture. The process of identifying and categorizing these diseases is often time-consuming and prone to errors. This research addresses this issue by employing a convolutional neural network and support vector machine (CNN-SVM) hybrid model to classify diseases in four economically important crops: strawberries, peaches, cherries, and soybeans. The objective is to categorize 10 classes of diseases, with six diseased classes and four healthy classes, for these crops using the deep learning-based CNN-SVM model. Several pre-trained models, including VGG16, VGG19, DenseNet, Inception, MobileNetV2, MobileNet, Xception, and ShuffleNet, were also trained, achieving accuracy ranges from 53.82% to 98.8%. The proposed model, however, achieved an average accuracy of 99.09%. While the proposed model's accuracy is comparable to that of the VGG16 pre-trained model, its significantly lower number of trainable parameters makes it more efficient and distinctive. This research demonstrates the potential of the CNN-SVM model in enhancing the accuracy and efficiency of plant disease classification. The CNN-SVM model was selected over VGG16 and other models due to its superior performance metrics. The proposed model achieved a 99% F1-score, a 99.98% Area Under the Curve (AUC), and a 99% precision value, demonstrating its efficacy. Additionally, class activation maps were generated using the Gradient Weighted Class Activation Mapping (Grad-CAM) technique to provide a visual explanation of the detected diseases. A heatmap was created to highlight the regions requiring classification, further validating the model's accuracy and interpretability.
Collapse
Affiliation(s)
- Reazul Hasan Prince
- Department of Electrical and Computer Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh
| | - Abdul Al Mamun
- Department of Computer Science and Engineering, Tejgaon College, Dhaka, Bangladesh
| | - Hasibul Islam Peyal
- Department of Electrical and Computer Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh
- Department of Computer Science and Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh
| | - Shafiun Miraz
- Department of Computer Science and Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh
| | - Md. Nahiduzzaman
- Department of Electrical and Computer Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh
| | - Amith Khandakar
- Department of Electrical Engineering, College of Engineering, Qatar University, Doha, Qatar
| | - Mohamed Arselene Ayari
- Department of Civil and Architectural Engineering, Qatar University, Doha, Qatar
- Technology Innovation and Engineering Education Unit, Qatar University, Doha, Qatar
| |
Collapse
|
2
|
Jamil M, Rehman H, Saqlain Zaheer M, Tariq A, Iqbal R, Hasnain MU, Majeed A, Munir A, Sabagh AE, Habib Ur Rahman M, Raza A, Ajmal Ali M, Elshikh MS. The use of Multispectral Radio-Meter (MSR5) data for wheat crop genotypes identification using machine learning models. Sci Rep 2023; 13:19867. [PMID: 37963968 PMCID: PMC10645743 DOI: 10.1038/s41598-023-46957-5] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2023] [Accepted: 11/07/2023] [Indexed: 11/16/2023] Open
Abstract
Satellite remote sensing is widely being used by the researchers and geospatial scientists due to its free data access for land observation and agricultural activities monitoring. The world is suffering from food shortages due to the dramatic increase in population and climate change. Various crop genotypes can survive in harsh climatic conditions and give more production with less disease infection. Remote sensing can play an essential role in crop genotype identification using computer vision. In many studies, different objects, crops, and land cover classification is done successfully, while crop genotypes classification is still a gray area. Despite the importance of genotype identification for production planning, a significant method has yet to be developed to detect the genotypes varieties of crop yield using multispectral radiometer data. In this study, three genotypes of wheat crop (Aas-'2011', 'Miraj-'08', and 'Punjnad-1) fields are prepared for the investigation of multispectral radio meter band properties. Temporal data (every 15 days from the height of 10 feet covering 5 feet in the circle in one scan) is collected using an efficient multispectral Radio Meter (MSR5 five bands). Two hundred yield samples of each wheat genotype are acquired and manually labeled accordingly for the training of supervised machine learning models. To find the strength of features (five bands), Principle Component Analysis (PCA), Linear Discriminant Analysis (LDA), and Nonlinear Discernment Analysis (NDA) are performed besides the machine learning models of the Extra Tree Classifier (ETC), Random Forest (RF), Support Vector Machine (SVM), Decision Tree (DT), Logistic Regression (LR), k Nearest Neighbor (KNN) and Artificial Neural Network (ANN) with detailed of configuration settings. ANN and random forest algorithm have achieved approximately maximum accuracy of 97% and 96% on the test dataset. It is recommended that digital policymakers from the agriculture department can use ANN and RF to identify the different genotypes at farmer's fields and research centers. These findings can be used for precision identification and management of the crop specific genotypes for optimized resource use efficiency.
Collapse
Affiliation(s)
- Mutiullah Jamil
- Department of Computer Science, Khwaja Fareed University of Engineering and Information Technology, Rahim Yar Khan, 64200, Pakistan
| | - Hafeezur Rehman
- Department of Computer Science, Khwaja Fareed University of Engineering and Information Technology, Rahim Yar Khan, 64200, Pakistan
| | - Muhammad Saqlain Zaheer
- Department of Agricultural Engineering, Khwaja Fareed University of Engineering and Information Technology, Rahim Yar Khan, Pakistan
| | - Aqil Tariq
- Department of Wildlife, Fisheries and Aquaculture, Mississippi State University, 775 Stone Boulevard, Mississippi State, MS, 39762-9690, USA.
- State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan University, Wuhan, 430079, China.
| | - Rashid Iqbal
- Department of Agronomy, Faculty of Agriculture and Environment, The Islamia University of Bahawalpur, Bahawalpur, Pakistan.
| | - Muhammad Usama Hasnain
- Institute of Plant Breeding and Biotechnology, MNS-University of Agriculture, Multan, Pakistan
| | - Asma Majeed
- Institute of Agro-Industry & Environment, The Islamia University of Bahawalpur, Bahawalpur, Pakistan
| | - Awais Munir
- Institute of Agro-Industry & Environment, The Islamia University of Bahawalpur, Bahawalpur, Pakistan
| | - Ayman El Sabagh
- Department of Agronomy, Faculty of Agriculture, Kafrelsheikh University, Kafr El-Shaikh, 33516, Egypt
- Department of Field Crops, Faculty of Agriculture, Siirt University, Siirt, Turkey
| | - Muhammad Habib Ur Rahman
- Institute of Plant Breeding and Biotechnology, MNS-University of Agriculture, Multan, Pakistan
- Crop Science, Institute of Crop Science and Resource Conservation (INRES), University of Bonn, 53115, Bonn, Germany
| | - Ahsan Raza
- Crop Science, Institute of Crop Science and Resource Conservation (INRES), University of Bonn, 53115, Bonn, Germany.
- Leibniz Centre for Agricultural Landscape Research (ZALF), Eberswalder Straße 84, 15374, Müncheberg, Germany.
| | - Mohammad Ajmal Ali
- Department of Botany and Microbiology, College of Science, King Saud University, 11451, Riyadh, Saudi Arabia
| | - Mohamed S Elshikh
- Department of Botany and Microbiology, College of Science, King Saud University, 11451, Riyadh, Saudi Arabia
| |
Collapse
|
3
|
Liu Y, Song Y, Ye R, Zhu S, Huang Y, Chen T, Zhou J, Li J, Li M, Lv C. High-Precision Tomato Disease Detection Using NanoSegmenter Based on Transformer and Lightweighting. PLANTS (BASEL, SWITZERLAND) 2023; 12:2559. [PMID: 37447120 DOI: 10.3390/plants12132559] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/28/2023] [Revised: 07/01/2023] [Accepted: 07/03/2023] [Indexed: 07/15/2023]
Abstract
With the rapid development of artificial intelligence and deep learning technologies, their applications in the field of agriculture, particularly in plant disease detection, have become increasingly extensive. This study focuses on the high-precision detection of tomato diseases, which is of paramount importance for agricultural economic benefits and food safety. To achieve this aim, a tomato disease image dataset was first constructed, and a NanoSegmenter model based on the Transformer structure was proposed. Additionally, lightweight technologies, such as the inverted bottleneck technique, quantization, and sparse attention mechanism, were introduced to optimize the model's performance and computational efficiency. The experimental results demonstrated excellent performance of the model in tomato disease detection tasks, achieving a precision of 0.98, a recall of 0.97, and an mIoU of 0.95, while the computational efficiency reached an inference speed of 37 FPS. In summary, this study provides an effective solution for high-precision detection of tomato diseases and offers insights and references for future research.
Collapse
Affiliation(s)
- Yufei Liu
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| | - Yihong Song
- College of Plant Protection, China Agricultural University, Beijing 100083, China
| | - Ran Ye
- College of Economics and Management, China Agricultural University, Beijing 100083, China
| | - Siqi Zhu
- College of Plant Protection, China Agricultural University, Beijing 100083, China
- International College Beijing, China Agricultural University, Beijing 100083, China
| | - Yiwen Huang
- College of Plant Protection, China Agricultural University, Beijing 100083, China
- International College Beijing, China Agricultural University, Beijing 100083, China
| | - Tailai Chen
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| | - Junyu Zhou
- College of Plant Protection, China Agricultural University, Beijing 100083, China
- International College Beijing, China Agricultural University, Beijing 100083, China
| | - Jiapeng Li
- School of Computer Science and Engineering, Beihang University, Beijing 100191, China
| | - Manzhou Li
- College of Plant Protection, China Agricultural University, Beijing 100083, China
| | - Chunli Lv
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| |
Collapse
|
4
|
Liu Y, Liu J, Cheng W, Chen Z, Zhou J, Cheng H, Lv C. A High-Precision Plant Disease Detection Method Based on a Dynamic Pruning Gate Friendly to Low-Computing Platforms. PLANTS (BASEL, SWITZERLAND) 2023; 12:plants12112073. [PMID: 37299053 DOI: 10.3390/plants12112073] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/03/2023] [Revised: 04/06/2023] [Accepted: 05/17/2023] [Indexed: 06/12/2023]
Abstract
Timely and accurate detection of plant diseases is a crucial research topic. A dynamic-pruning-based method for automatic detection of plant diseases in low-computing situations is proposed. The main contributions of this research work include the following: (1) the collection of datasets for four crops with a total of 12 diseases over a three-year history; (2) the proposition of a re-parameterization method to improve the boosting accuracy of convolutional neural networks; (3) the introduction of a dynamic pruning gate to dynamically control the network structure, enabling operation on hardware platforms with widely varying computational power; (4) the implementation of the theoretical model based on this paper and the development of the associated application. Experimental results demonstrate that the model can run on various computing platforms, including high-performance GPU platforms and low-power mobile terminal platforms, with an inference speed of 58 FPS, outperforming other mainstream models. In terms of model accuracy, subclasses with a low detection accuracy are enhanced through data augmentation and validated by ablation experiments. The model ultimately achieves an accuracy of 0.94.
Collapse
Affiliation(s)
- Yufei Liu
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| | - Jingxin Liu
- College of Economics and Management, China Agricultural University, Beijing 100083, China
| | - Wei Cheng
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| | - Zizhi Chen
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| | - Junyu Zhou
- International College Beijing, China Agricultural University, Beijing 100083, China
| | - Haolan Cheng
- International College Beijing, China Agricultural University, Beijing 100083, China
| | - Chunli Lv
- College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China
| |
Collapse
|
5
|
Tariq A, Mumtaz F, Majeed M, Zeng X. Spatio-temporal assessment of land use land cover based on trajectories and cellular automata Markov modelling and its impact on land surface temperature of Lahore district Pakistan. ENVIRONMENTAL MONITORING AND ASSESSMENT 2022; 195:114. [PMID: 36385403 DOI: 10.1007/s10661-022-10738-w] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/07/2022] [Accepted: 11/05/2022] [Indexed: 06/16/2023]
Abstract
This research aims to assess the urban growth and impact on land surface temperature (LST) of Lahore, the second biggest city in Pakistan. In this research, various geographical information system (GIS) and remote sensing (RS) techniques (maximum likelihood classification (MLC)) LST, and different normalized satellite indices have been implemented to analyse the spatio-temporal trends of Lahore city; by using Landsat for 1990, 2004, and 2018. The development of integrated use of RS and GIS and combined cellular automata-Markov models has provided new means of assessing changes in land use and land cover and has enabled the projection of trajectories into the future. Results indicate that the built-up area and bare area increased from 15,541 (27%) to 23,024 km2 (40%) and 5756 km2 (10%) to 13,814 km2 (24%). Meanwhile, water area and vegetation were decreased from 2302 km2 (4%) to 1151 km2 (2%) and 33,961 km2 (59%) to 19,571 km2 (34%) respectively. Under this urbanization, the LST of the city was also got affected. In 1990, the mean LST of most of the area was between 14 and 28 ℃, which rose to 22-28 ℃ in 2004 and 34 to 36 ℃ in 2018. Because of the shift of vegetation and built-up land, the surface reflectance and roughness of each land use land cover (LULC) class are different. The analysis established a direct correlation between Normalized Difference Water Index (NDWI), Normalized Difference Vegetation Index (NDVI) with LST and an indirect correlation among Soil Adjusted Vegetation Index (SAVI), Normalized Difference Built-up Index (NDBI), and Built-up Index (BI) with LST. The results are important for the planning and development department since they may be used to guarantee the sustainable utilization of land resources for future urbanization expansion projects.
Collapse
Affiliation(s)
- Aqil Tariq
- Department of Wildlife, Fisheries and Aquaculture, College of Forest Resources, Mississippi State University, 775 Stone Boulevard, Mississippi State, MS, 39762-9690, USA.
- State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan University, Wuhan, 430079, China.
| | - Faisal Mumtaz
- University of Chinese Academy of Sciences (UCAS), Beijing, 101408, China
- State Key Laboratory of Remote Sensing Sciences, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, 100101, China
| | - Muhammad Majeed
- Department of Botany, University of Gujrat, Hafiz Hayat Campus, Gujrat, Punjab, Pakistan
| | - Xing Zeng
- State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing (LIESMARS), Wuhan University, Wuhan, 430079, China
| |
Collapse
|
6
|
Pan J, Xia L, Wu Q, Guo Y, Chen Y, Tian X. Automatic strawberry leaf scorch severity estimation via faster R-CNN and few-shot learning. ECOL INFORM 2022. [DOI: 10.1016/j.ecoinf.2022.101706] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
|
7
|
Niu Q, Liu J, Jin Y, Chen X, Zhu W, Yuan Q. Tobacco shred varieties classification using Multi-Scale-X-ResNet network and machine vision. FRONTIERS IN PLANT SCIENCE 2022; 13:962664. [PMID: 36061766 PMCID: PMC9433752 DOI: 10.3389/fpls.2022.962664] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/06/2022] [Accepted: 07/25/2022] [Indexed: 05/21/2023]
Abstract
The primary task in calculating the tobacco shred blending ratio is identifying the four tobacco shred types: expanded tobacco silk, cut stem, tobacco silk, and reconstituted tobacco shred. The classification precision directly affects the subsequent determination of tobacco shred components. However, the tobacco shred types, especially expanded tobacco silk and tobacco silk, have no apparent differences in macro-scale characteristics. The tobacco shreds have small size and irregular shape characteristics, creating significant challenges in their recognition and classification based on machine vision. This study provides a complete set of solutions aimed at this problem for screening tobacco shred samples, taking images, image preprocessing, establishing datasets, and identifying types. A block threshold binarization method is used for image preprocessing. Parameter setting and method performance are researched to obtain the maximum number of complete samples with acceptable execution time. ResNet50 is used as the primary classification and recognition network structure. By increasing the multi-scale structure and optimizing the number of blocks and loss function, a new tobacco shred image classification method is proposed based on the MS-X-ResNet (Multi-Scale-X-ResNet) network. Specifically, the MS-ResNet network is obtained by fusing the multi-scale Stage 3 low-dimensional and Stage 4 high-dimensional features to reduce the overfitting risk. The number of blocks in Stages 1-4 are adjusted from the original 3:4:6:3 to 3:4:N:3 (A-ResNet) and 3:3:N:3 (B-ResNet) to obtain the X-ResNet network, which improves the model's classification performance with lower complexity. The focal loss function is selected to reduce the impact of identification difficulty for different sample types on the network and improve its performance. The experimental results show that the final classification accuracy of the network on a tobacco shred dataset is 96.56%. The image recognition of a single tobacco shred requires 103 ms, achieving high classification accuracy and efficiency. The image preprocessing and deep learning algorithms for tobacco shred classification and identification proposed in this study provide a new implementation approach for the actual production and quality detection of tobacco and a new way for online real-time type identification of other agricultural products.
Collapse
Affiliation(s)
- Qunfeng Niu
- School of Electrical Engineering, Henan University of Technology, Zhengzhou, China
| | - Jiangpeng Liu
- School of Electrical Engineering, Henan University of Technology, Zhengzhou, China
| | - Yi Jin
- Anyang Cigarette Factory, China Tobacco Henan Industrial Co., Ltd., Anyang, China
| | - Xia Chen
- Anyang Cigarette Factory, China Tobacco Henan Industrial Co., Ltd., Anyang, China
| | - Wenkui Zhu
- Zhengzhou Tobacco Research Institute of China National Tobacco Corporation (CNTC), Zhengzhou, China
| | - Qiang Yuan
- School of Electrical Engineering, Henan University of Technology, Zhengzhou, China
| |
Collapse
|
8
|
Albattah W, Javed A, Nawaz M, Masood M, Albahli S. Artificial Intelligence-Based Drone System for Multiclass Plant Disease Detection Using an Improved Efficient Convolutional Neural Network. FRONTIERS IN PLANT SCIENCE 2022; 13:808380. [PMID: 35755664 PMCID: PMC9218756 DOI: 10.3389/fpls.2022.808380] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/04/2021] [Accepted: 04/08/2022] [Indexed: 05/31/2023]
Abstract
The role of agricultural development is very important in the economy of a country. However, the occurrence of several plant diseases is a major hindrance to the growth rate and quality of crops. The exact determination and categorization of crop leaf diseases is a complex and time-required activity due to the occurrence of low contrast information in the input samples. Moreover, the alterations in the size, location, structure of crop diseased portion, and existence of noise and blurriness effect in the input images further complicate the classification task. To solve the problems of existing techniques, a robust drone-based deep learning approach is proposed. More specifically, we have introduced an improved EfficientNetV2-B4 with additional added dense layers at the end of the architecture. The customized EfficientNetV2-B4 calculates the deep key points and classifies them in their related classes by utilizing an end-to-end training architecture. For performance evaluation, a standard dataset, namely, the PlantVillage Kaggle along with the samples captured using a drone is used which is complicated in the aspect of varying image samples with diverse image capturing conditions. We attained the average precision, recall, and accuracy values of 99.63, 99.93, and 99.99%, respectively. The obtained results confirm the robustness of our approach in comparison to other recent techniques and also show less time complexity.
Collapse
Affiliation(s)
- Waleed Albattah
- Department of Information Technology, College of Computer, Qassim University, Buraydah, Saudi Arabia
| | - Ali Javed
- Department of Computer Science, University of Engineering and Technology Taxila, Taxila, Pakistan
| | - Marriam Nawaz
- Department of Computer Science, University of Engineering and Technology Taxila, Taxila, Pakistan
| | - Momina Masood
- Department of Computer Science, University of Engineering and Technology Taxila, Taxila, Pakistan
| | - Saleh Albahli
- Department of Information Technology, College of Computer, Qassim University, Buraydah, Saudi Arabia
| |
Collapse
|