Paddy disease detection system in agriculture, has model for learning patterns associated with paddy diseases effectively and final layers of network retrained while maintaining pre-trained weights for feature extraction
	HABIBA U H; VASUMATHI G; BALAJI B V; REDDY V V A; REDDY G A; NIKHIL B
	2023-10-16
专利权人	BHARATH INST HIGHER EDUCATION & RES (BHAR-Non-standard)
申请日期	2023-10-16
专利号	IN202341070033-A
成果简介	NOVELTY - The system has paddy leaf disease data which extracts and classifies feature using 50 layer residual network (ResNet 50) and visual geometry group with sixteen layers depth (VGG16) model with continuous data collection and model updates to maintain effectiveness in disease recognition. The images are pre-processed properly and are resized, normalized and augmented. A model learns patterns associated with paddy diseases effectively. The VGG16 is a powerful pre-trained model and is fine-tuned on a paddy disease dataset to adapt to specific task. Final layers of a network are retrained while maintaining pre-trained weights for feature extraction. The performance of the model is rigorously evaluated using appropriate metrics e.g. accuracy, precision, recall, F1-score and confusion matrices, to determine effectiveness in disease recognition of paddy disease images. The model is trained on a diverse set of images that encompass different types and stages of diseases to generalize well. USE - Paddy disease detection system in agriculture. ADVANTAGE - The system reduces time-consuming to identify paddy leaf diseases due to the unavailability of experts, avoids incorrect actions by experts, suggests the proper treatment to cure disease to consider appropriate actions within the time, improves accuracy, scalability, integration and interpretability, detects paddy leaf disease, provides regular updates, interprets model interpretability e.g. gradient-based visualization and saliency maps and helps the farmers to know the treatment of the diseased plant.
IPC 分类号	G06N-003/04 ; G06N-003/08 ; G06N-007/00 ; G06Q-050/02 ; G16B-020/00
国家	印度
专业领域	信息技术
语种	英语
成果类型	专利
文献类型	科技成果
条目标识符	http://119.78.100.226:8889/handle/3KE4DYBR/19860
专题	中国科学院新疆生态与地理研究所
作者单位	BHARATH INST HIGHER EDUCATION & RES (BHAR-Non-standard)
推荐引用方式 GB/T 7714	HABIBA U H,VASUMATHI G,BALAJI B V,et al. Paddy disease detection system in agriculture, has model for learning patterns associated with paddy diseases effectively and final layers of network retrained while maintaining pre-trained weights for feature extraction. IN202341070033-A[P]. 2023.

条目包含的文件
条目无相关文件。

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。