ESFD-YOLOv8n: Early Smoke and Fire Detection Method Based on an Improved YOLOv8n Model

Fire, Год журнала: 2024, Номер 7(9), С. 303 - 303

Опубликована: Авг. 27, 2024

Ensuring fire safety is essential to protect life and property, but modern infrastructure complex settings require advanced detection methods. Traditional object systems, often reliant on manual feature extraction, may fall short, while deep learning approaches are powerful, they can be computationally intensive, especially for real-time applications. This paper proposes a novel smoke method based the YOLOv8n model with several key architectural modifications. The standard Complete-IoU (CIoU) box loss function replaced more robust Wise-IoU version 3 (WIoUv3), enhancing predictions through its attention mechanism dynamic focusing. streamlined by replacing C2f module residual block, enabling targeted accelerating training inference, reducing overfitting. Integrating generalized efficient layer aggregation network (GELAN) blocks modules in neck of further enhances detection, optimizing gradient paths high performance. Transfer also applied enhance robustness. Experiments confirmed excellent performance ESFD-YOLOv8n, outperforming original 2%, 2.3%, 2.7%, mean average precision ([email protected]) 79.4%, 80.1%, recall 72.7%. Despite increased complexity, outperforms state-of-the-art algorithms meets requirements detection.

Язык: Английский

---

Research on an Apple Recognition and Yield Estimation Model Based on the Fusion of Improved YOLOv11 and DeepSORT

Agriculture, Год журнала: 2025, Номер 15(7), С. 765 - 765

Опубликована: Апрель 2, 2025

Accurate apple yield estimation is essential for effective orchard management, market planning, and ensuring growers’ income. However, complex conditions, such as dense foliage occlusion overlapping fruits, present challenges to large-scale estimation. This study introduces APYOLO, an enhanced detection algorithm based on improved YOLOv11, integrated with the DeepSORT tracking improve both accuracy operational speed. APYOLO incorporates a multi-scale channel attention (MSCA) mechanism prior distribution intersection over union (EnMPDIoU) loss function enhance target localization recognition under environments. Experimental results demonstrate that outperforms original YOLOv11 by improving [email protected], [email protected]–0.95, accuracy, recall 2.2%, 2.1%, 0.8%, 2.3%, respectively. Additionally, combination of unique ID region line (ROL) strategy in further boosts 84.45%, surpassing performance method alone. provides more precise efficient system estimation, offering strong technical support intelligent refined management.

Язык: Английский

---

YOLO-STOD: an industrial conveyor belt tear detection model based on Yolov5 algorithm

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 11, 2025

Real-time detection of conveyor belt tearing is great significance to ensure mining in the coal industry. The longitudinal tear damage problem belts has characteristics multi-scale, abundant small targets, and complex interference sources. Therefore, order improve performance small-size algorithms under interference, a visual method YOLO-STOD based on deep learning was proposed. Firstly, multi-case dataset developed for detection. Second, designed, which utilizes BotNet attention mechanism extract multi-dimensional features, enhancing model's feature extraction ability targets enables model converge quickly conditions few samples. Secondly, Shape_IOU utilized calculate training loss, shape regression loss bounding box itself considered enhance robustness model. experimental results fully proved effectiveness method, constantly surpasses competing methods achieves 91.2%, 91.9%, 190.966 accuracy speed terms recall, Map value, FPS, respectively, able satisfy needs industrial real-time expected be used field.

Язык: Английский

---

YOLOv8-CBSE: An Enhanced Computer Vision Model for Detecting the Maturity of Chili Pepper in the Natural Environment

Agronomy, Год журнала: 2025, Номер 15(3), С. 537 - 537

Опубликована: Фев. 23, 2025

In order to accurately detect the maturity of chili peppers under different lighting and natural environmental scenarios, in this study, we propose a lightweight detection model, YOLOv8-CBSE, based on YOLOv8n. By replacing C2f module original model with designed C2CF module, integrates advantages convolutional neural networks Transformer architecture, improving model’s ability extract local features global information. Additionally, SRFD DRFD modules are introduced replace layers, effectively capturing at scales enhancing diversity adaptability through feature fusion mechanism. To further improve accuracy, EIoU loss function is used instead CIoU provide more comprehensive The results showed that average precision (AP) YOLOv8-CBSE for mature immature was 90.75% 85.41%, respectively, F1 scores mean (mAP) 81.69% 88.08%, respectively. Compared YOLOv8n, score mAP improved increased by 0.46% 1.16%, effect pepper scenarios improved, which proves robustness YOLOv8-CBSE. also maintains design size only 5.82 MB, its suitability real-time applications resource-constrained devices. This study provides an efficient accurate method detecting environments, great significance promoting intelligent precise agricultural management.

Язык: Английский

---

A Novel Fusion Perception Algorithm of Tree Branch/Trunk and Apple for Harvesting Robot Based on Improved YOLOv8s

Agronomy, Год журнала: 2024, Номер 14(9), С. 1895 - 1895

Опубликована: Авг. 24, 2024

Aiming to accurately identify apple targets and achieve segmentation the extraction of branch trunk areas trees, providing visual guidance for a picking robot actively adjust its posture avoid trunks obstacle avoidance fruit picking, spindle-shaped which are widely planted in standard modern orchards, were focused on, an algorithm tree detection robots was proposed based on improved YOLOv8s model design. Firstly, image data trees orchards collected, annotations object pixel-level conducted data. Training set then augmented improve generalization performance algorithm. Secondly, original network architecture’s design by embedding SE module attention mechanism after C2f Backbone architecture. Finally, dynamic snake convolution embedded into Neck structure architecture better extract feature information different branches. The experimental results showed that can effectively recognize images segment branches trunks. For recognition, precision 99.6%, recall 96.8%, mAP value 98.3%. 81.6%. compared with YOLOv8s, YOLOv8n, YOLOv5s algorithms recognition test images. other three algorithms, increased 1.5%, 2.3%, 6%, respectively. 3.7%, 15.4%, 24.4%, fruits, branches, is great significance ensuring success rate harvesting, provide technical support development intelligent harvesting robot.

Язык: Английский

---

A case study on artifical intelligence based data processing in passive brain–computer interface

Elsevier eBooks, Год журнала: 2025, Номер unknown, С. 227 - 250

Опубликована: Янв. 1, 2025

Язык: Английский

---

A Comprehensive Review of Deep Learning in Computer Vision for Monitoring Apple Tree Growth and Fruit Production

Sensors, Год журнала: 2025, Номер 25(8), С. 2433 - 2433

Опубликована: Апрель 12, 2025

The high nutritional and medicinal value of apples has contributed to their widespread cultivation worldwide. Unfavorable factors in the healthy growth trees extensive orchard work are threatening profitability apples. This study reviewed deep learning combined with computer vision for monitoring apple tree fruit production processes past seven years. Three types models were used real-time target recognition tasks: detection including You Only Look Once (YOLO) faster region-based convolutional network (Faster R-CNN); classification Alex (AlexNet) residual (ResNet); segmentation (SegNet), mask regional neural (Mask R-CNN). These have been successfully applied detect pests diseases (located on leaves, fruits, trunks), organ (including blossoms, branches), yield, post-harvest defects. introduced methods, outlined current research these methods production. advantages disadvantages discussed, difficulties faced future trends summarized. It is believed that this important construction smart orchards.

Язык: Английский

---

PSR-LeafNet: A Deep Learning Framework for Identifying Medicinal Plant Leaves Using Support Vector Machines

Big Data and Cognitive Computing, Год журнала: 2024, Номер 8(12), С. 176 - 176

Опубликована: Дек. 1, 2024

In computer vision, recognizing plant pictures has emerged as a multidisciplinary area of interest. the last several years, much research been conducted to determine type in each image automatically. The challenges identifying medicinal plants are due changes effects light, stance, and orientation. Further, it is difficult identify factors like variations leaf shape with age changing color response varying weather conditions. proposed work uses machine learning techniques deep neural networks choose appropriate features if or non-medicinal plant. This study presents network design based on PSR-LeafNet (PSR-LN). single that combines P-Net, S-Net, R-Net, all intended for feature extraction using minimum redundancy maximum relevance (MRMR) approach. PSR-LN helps obtain features, venation leaf, textural features. A support vector (SVM) applied output achieved from PSR network, which classify name model named PSR-LN-SVM. advantage designed suits more considerable dataset processing provides better results than traditional models. methodology utilized achieves an accuracy 97.12% MalayaKew dataset, 98.10% IMP 95.88% Flavia dataset. models surpass existing models, having improvement accuracy. These outcomes demonstrate suggested method successful accurately leaves plants, paving way advanced taxonomy medicine.

Язык: Английский

---

A Study of Kale Recognition Based on Semantic Segmentation

Agronomy, Год журнала: 2024, Номер 14(5), С. 894 - 894

Опубликована: Апрель 25, 2024

The kale crop is an important bulk vegetable, and automatic segmentation to recognize fundamental for effective field management. However, complex backgrounds texture-rich edge details make fine of difficult. To this end, we constructed a dataset in real scenario proposed UperNet semantic model with Swin transformer as the backbone network improved according growth characteristics kale. Firstly, channel attention module (CAM) introduced into improve representation ability enhance extraction outer leaf bulb information; secondly, accuracy target edges decoding part by designing refinement (ARM); lastly, uneven distribution classes solved modifying optimizer loss function solve class problem. experimental results show that paper has excellent performance feature extraction, average intersection merger ratio (mIOU) can be up 91.2%, pixel (mPA) 95.2%, which 2.1 percentage points 4.7 higher than original model, respectively, it effectively improves recognition

Язык: Английский

---

RGB-D Camera and Fractal-Geometry-Based Maximum Diameter Estimation Method of Apples for Robot Intelligent Selective Graded Harvesting

Fractal and Fractional, Год журнала: 2024, Номер 8(11), С. 649 - 649

Опубликована: Ноя. 7, 2024

Realizing the integration of intelligent fruit picking and grading for apple harvesting robots is an inevitable requirement future development smart agriculture precision agriculture. Therefore, maximum diameter estimation model based on RGB-D camera fusion depth information was proposed in study. Firstly, parameters Red Fuji apples were collected, results statistically analyzed. Then, Intel RealSense D435 LabelImg software, two-dimensional size images obtained. Furthermore, relationship between information, images, explored. Based Origin multiple regression analysis nonlinear surface fitting used to analyze correlation depth, diagonal length bounding rectangle, diameter. A estimating constructed. Finally, constructed experimentally validated evaluated imitation laboratory fruits trees modern orchards. The experimental showed that average relative error validation set ±4.1%, coefficient (R2) estimated 0.98613, root mean square (RMSE) 3.21 mm. orchard ±3.77%, 0.84, 3.95 can provide theoretical basis technical support selective apple-picking operation grading.

Язык: Английский

---