2024 · 12 citations · 13 references
ComputingComputer SciencePrecision AgricultureComputer VisionMachine LearningPlant Disease DiagnosisAgricultural Machine VisionDeep LearningDisease DetectionImage ClassificationRepresentation LearningTransfer LearningBacterial Leaf SpotCrop Disease ClassificationDeep Learning DetectionImage Classification For CropsLeaf CurlLeaf Spot DiseasesLeaf Spot SymptomsLeaf Symptom Recognition
The cotton leaf disease infliction anywhere puts the production of agricultural products at risk as well as sustainable farming causing both problems and solutions of the meaningless disease detection. In this research, we propose an optimized Visual Geometry Group (VGG) 16 model for the advanced classification of four major cotton leaf diseases: leaf curl, bacterial blighty, leaf spot, and wilt. Integration of deep learning methods, transfer learning as well and other approaches is what the model focuses on to raise the disease detection accuracy rate and realize the management of agriculture precision. And, there was a complete dataset, which included images of cotton plants from infected leaves with various diseases. This dataset was annotated properly and preprocessed. VGG16 model was pre-trained with a substantial data set of ImageNet image database and then fine-tuned to perform the cotton leaf disease classification task using existing transfer learning techniques. The achieved model efficacy has been evaluated using commonly accepted evaluation metrics. These include accuracy, precision, recall, and F1-score. Results show that the invented technique performs well when the revised VGG 16 model obtains satisfactory accuracy and disease-related metrics when applied to all three disease categories. The level of comparative analysis with current best models justifies done method efficacy and its strength in cotton leaf disease classification. This study’s findings have an essential consequence on precision farming. It grants farmers a scalable solution relevant for the quick and automated monitoring, early diagnosis, and possibly treatment of plant diseases. Future investigation may rule in the superiority of applying the modified model to practical agricultural fields for ready accessibility and usability. Ultimately, these findings are moving us into the phase of development of agricultural informatics and offer evidence of the ability of deep learning to radically change approaches to fighting crop disease.
13
Deep learning for identifying radiogenomic associations in breast cancer
Zhe Zhu, Ehab A. AlBadawy, Ashirbani Saha et al. · Computers in Biology and Medicine · 2019 · 163 citations · Full text
Medicine, Bioinformatics, Oncology +15