Article
Plant Sciences
Jia-Rong Xiao, Pei-Che Chung, Hung-Yi Wu, Quoc-Hung Phan, Jer-Liang Andrew Yeh, Max Ti-Kuang Hou
Summary: Strawberry is a high-value crop in Taiwan, but various diseases have significantly decreased production. Anthracnose crown rot caused substantial losses from 2010 to 2016. Utilizing a convolutional neural network model for image recognition can effectively detect strawberry diseases.
Article
Agriculture, Multidisciplinary
Dongfang Wang, Jun Wang, Wenrui Li, Ping Guan
Summary: This paper discusses the importance of detecting plant diseases and the application of convolutional neural networks in crop and disease recognition, proposing a new method that separates crop and disease identification and demonstrating its effectiveness. The results show high accuracy in crop and disease identification in controlled laboratory environments, and decent accuracy in real-world environments as well.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Computer Science, Information Systems
Ronald Wihal Oei, Wynne Hsu, Mong Li Lee, Ngiap Chuan Tan
Summary: This study develops a convolutional neural network-based learning framework for retrieving clinically similar patients and personalizing the prediction of macrovascular complications. The results show that this framework effectively retrieves similar patients and outperforms other approaches in complication prediction. It provides guidance for physicians in choosing effective treatments.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Computer Science, Artificial Intelligence
Ali Ghofrani, Rahil Mahdian Toroghi
Summary: This paper proposes a deep learning approach for recognizing plant diseases during their critical period. A client-server system is designed to consider both performance and accuracy, with a novel knowledge distillation technique improving the accuracy of the small client-side model. The experimental results show a significant improvement in the classification rate of the state-of-the-art tiny model by using this teacher-student idea.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Plant Sciences
Irfan Abbas, Jizhan Liu, Muhammad Amin, Aqil Tariq, Mazhar Hussain Tunio
Summary: Plant health is crucial for agricultural development, and diseases significantly impact crop losses. Deep learning models, such as EfficientNet-B3 and VGG-16, show high accuracy in identifying leaf scorch disease in strawberry plants. These trained CNN models can be integrated with a machine vision system for real-time image acquisition and disease identification.
Article
Computer Science, Artificial Intelligence
Jacson Rodrigues Correia-Silva, Rodrigo F. Berriel, Claudine Badue, Alberto F. De Souza, Thiago Oliveira-Santos
Summary: Convolutional neural networks have been successful in enabling companies to develop neural-based products, but concerns arise about model security and copying. The authors proposed a method to copy black-box models and further consolidated and expanded the approach, demonstrating the effectiveness of natural random images in generating copycats for various problems.
PATTERN RECOGNITION
(2021)
Article
Mathematics, Interdisciplinary Applications
Li Ma, Xueliang Guo, Shuke Zhao, Doudou Yin, Yiyi Fu, Peiqi Duan, Bingbing Wang, Li Zhang
Summary: This study explores the recognition of common strawberry diseases using deep convolution neural network technology and proposes a new strawberry disease recognition algorithm. By training and transfer learning, the efficiency and accuracy of disease recognition are improved through effective identification and classification of strawberry diseases.
Article
Multidisciplinary Sciences
Ruiqing Wang, Wu Zhang, Jiuyang Ding, Meng Xia, Mengjian Wang, Yuan Rao, Zhaohui Jiang
Summary: The paper proposed a DNN-based compression method to reduce computational burden and compress model size through lightweight fully connected layers, pruning, knowledge distillation, and quantization. The experiment demonstrated that the compressed model can be reduced to 0.04 Mb with an accuracy of 97.09%, proving the effectiveness of knowledge distillation and the efficiency of compressed models over prevalent lightweight models.
Review
Biochemical Research Methods
Jun Liu, Xuewei Wang
Summary: Deep learning technology has made significant progress in plant disease and pest identification, showing advantages over traditional methods. A review of recent research based on deep learning outlines the classification, detection, and segmentation networks used for plant disease and pest detection, summarizing the pros and cons of each method.
Article
Computer Science, Artificial Intelligence
Zhiyong Huang, Kekai Sheng, Ke Li, Jian Liang, Taiping Yao, Weiming Dong, Dengwen Zhou, Xing Sun
Summary: Batch normalization is widely used in deep neural networks, but it is ineffective for cross-domain tasks. This paper proposes a novel normalization method called Reciprocal Normalization (RN), which utilizes cross-domain relation to improve adaptability. Compared to batch normalization, RN is more suitable for unsupervised domain adaptation and can be easily integrated into popular domain adaptation methods.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Information Systems
Keji Han, Yun Li, Bin Xia
Summary: This study introduces a new adversarial example detection method CMAG, which combines model-aware and generative technologies to address the issue of adversarial examples effectively and in an interpretable manner compared to state-of-the-art methods.
TSINGHUA SCIENCE AND TECHNOLOGY
(2021)
Article
Computer Science, Software Engineering
Zheng Zhan, Wenping Wang, Falai Chen
Summary: This article proposes a learning based method using a deep neural network to simultaneously parameterize the boundary and interior of a computational domain. The method achieves robust parameterization by optimizing a loss function and fitting a tensor-product B-spline function. Experimental results demonstrate that the proposed approach yields parameterization results with lower distortion and higher bijectivity ratio.
COMPUTER-AIDED DESIGN
(2024)
Article
Agriculture, Multidisciplinary
Jaemyung Shin, Young K. Chang, Brandon Heung, Tri Nguyen-Quang, Gordon W. Price, Ahmad Al-Mallahi
Summary: This study utilized Deep Learning models to detect powdery mildew in strawberries, optimizing well-established learners and performing data augmentation to prevent overfitting. The six DL algorithms used showed an average classification accuracy of over 92%, with ResNet-50 performing the best in classification.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Environmental Sciences
Jiangbo Xi, Okan K. Ersoy, Ming Cong, Chaoying Zhao, Wei Qu, Tianjun Wu
Summary: This paper proposes a wide and deep Fourier network for efficient feature learning in hyperspectral remote sensing image (HSI) classification. The method utilizes pruned features extracted in the frequency domain to extract hierarchical features layer-by-layer. Experimental results show that the proposed method achieves excellent performance in HSI classification, with the ability to be implemented in lightweight embedded computing platforms.
Article
Automation & Control Systems
Wenkai Yang, Yunyun Dong, Qianqian Du, Yan Qiang, Kun Wu, Juanjuan Zhao, Xiaotang Yang, Muhammad Bilal Zia
Summary: The study proposes a multi-task cascade deep learning model that integrates radiologists' domain knowledge and uses multimodal ultrasound images for automatic thyroid nodule diagnosis. Experimental results show that the model can achieve similar classification performance to fully supervised learning with only about 35% of labeled dataset, saving time and effort compared to traditional methods.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2021)