Plant pest and disease lightweight identification model by fusing tensor features and knowledge distillation

Frontiers in Plant Science, Год журнала: 2024, Номер 15

Опубликована: Ноя. 21, 2024

Plant pest and disease management is an important factor affecting the yield quality of crops, due to rich variety diagnosis process mostly relying on experts' experience, there are problems low efficiency accuracy. For this, we proposed a Disease Lightweight identification Model by fusing Tensor features Knowledge distillation (PDLM-TK). First, Residual Blocks based Spatial (LRB-ST) constructed enhance perception extraction shallow detail plant images introducing spatial tensor. And depth separable convolution used reduce number model parameters improve efficiency. Secondly, Branch Network Fusion with Graph Convolutional (BNF-GC) realize image super-pixel segmentation using spanning tree clustering pixel features. graph neural network utilized extract correlation Finally, designed Training Strategy knowledge Distillation (MTS-KD) train building migration architecture, which fully balances accuracy model. The experimental results show that PDLM-TK performs well in three datasets such as Village, highest classification F1 score 96.19% 94.94%. Moreover, execution better compared lightweight methods MobileViT, can quickly accurately diagnose diseases.

Язык: Английский

---

Modified ensemble machine learning-based plant leaf disease detection model with optimized K-Means clustering

Network Computation in Neural Systems, Год журнала: 2024, Номер unknown, С. 1 - 45

Опубликована: Дек. 9, 2024

In the farming sector, automatic detection of plant leaf disease is considered a vital landmark. Farmers move long distances to consult pathologists observe disease, which expensive and time-consuming. Moreover, in premature period difficult process existing model. Thus, all these challenges motivate us develop an inventive developed model, data gathered initially given as input pre-processing step using Contrast Limited Adaptive Histogram Equalization (CLAHE). Next, leaves are segmented from pre-processed images, then abnormality segmentation done by K-means clustering system. Here, parameters optimized Opposition-based Bird Swarm Algorithm (O-BSA). Further, features were extracted abnormality-segmented images feature extraction. The classification step, where carried out Optimized Ensemble Machine Learning (OEML), where, parameter optimization O-BSA. Finally, approach evaluated with various performance metrics, accuracy up 92.26. These findings show that model promising over conventional methods its effectiveness detecting disease.

Язык: Английский

---

A Real-Time Tracking System for Bread Production Based on YOLOv8 and DeepSORT

Lecture notes in networks and systems, Год журнала: 2024, Номер unknown, С. 469 - 479

Опубликована: Янв. 1, 2024

Язык: Английский

---

Big Data Visualization for Black Sigatoka Disease of Bananas and Pathogen–Host Interactions (PHI) of Other Plants

International Journal of Applied Research in Bioinformatics, Год журнала: 2024, Номер 13(1), С. 1 - 22

Опубликована: Ноя. 30, 2024

Black sigatoka is a leaf spot disease affecting banana plants that has caused significant yield reductions of up to 50% (Arman et al., 2023). This research presents data visualizations 8,761 points related black in plants, encompassing attributes such as time, canopy temperature, and relative humidity. The paper also reviews work, including the application mining plant studies, use deep learning neural networks for data, machine predicting crop yields detecting disease. Additionally, it big 20,952 values obtained from web-accessible Pathogen–Host Interactions database (PHI-base), covering various categories providing an epidemiological analysis prevalent causative agents specific diseases.

Язык: Английский

---

Mob-psp: modified MobileNet-V2 network for real-time detection of tomato diseases

Journal of Real-Time Image Processing, Год журнала: 2024, Номер 21(5)

Опубликована: Окт. 1, 2024

Язык: Английский

---

YOLOv5s-ECCW: A Lightweight Detection Model for Sugarcane Smut in Natural Environments

Agronomy, Год журнала: 2024, Номер 14(10), С. 2327 - 2327

Опубликована: Окт. 10, 2024

Sugarcane smut, a serious disease caused by the fungus Sporosorium scitamineum, can result in 30% to 100% cane loss. The most affordable and efficient measure of preventing handling sugarcane smut is select disease-resistant varieties. A comprehensive evaluation resistance based on incidence essential during selection process, necessitating rapid accurate identification smut. Traditional methods, which rely visual observation symptoms, are time-consuming, costly, inefficient. To address these limitations, we present lightweight detection model (YOLOv5s-ECCW), incorporates several innovative features. Specifically, EfficientNetV2 incorporated into YOLOv5 network achieve compression while maintaining high accuracy. convolutional block attention mechanism (CBAM) added backbone improve its feature extraction capability suppress irrelevant information. C3STR module used replace C3 module, enhancing ability capture global large targets. WIoU loss function place CIoU one bounding box regression’s experimental results demonstrate that YOLOv5s-ECCW achieves mean average precision (mAP) 97.8% with only 4.9 G FLOPs 3.25 M parameters. Compared original YOLOv5, our improvements include 0.2% increase mAP, 54% reduction parameters, 70.3% decrease computational requirements. proposed outperforms YOLOv4, SSD, YOLOv8 terms accuracy, efficiency, size. meets urgent need for real-time supporting better management resistant

Язык: Английский

---