Search Results (17)

Due to the limited computational resources of portable devices, target detection models for drone detection face challenges in real-time deployment. To enhance the detection efficiency of low, slow, and small unmanned aerial vehicles (UAVs), this study introduces an efficient drone detection model based on YOLOv5s (EDU-YOLO), incorporating lightweight feature extraction and balanced feature fusion modules. The model employs the ShuffleNetV2 network and coordinate attention mechanisms to construct a lightweight backbone network, significantly reducing the number of model parameters. It also utilizes a bidirectional feature pyramid network and ghost convolutions to build a balanced neck network, enriching the model’s representational capacity. Additionally, a new loss function, EloU, replaces CIoU to improve the model’s positioning accuracy and accelerate network convergence. Experimental results indicate that, compared to the YOLOv5s algorithm, our model only experiences a minimal decrease in mAP by 1.1%, while reducing GFLOPs from 16.0 to 2.2 and increasing FPS from 153 to 188. This provides a substantial foundation for networked optoelectronic detection of UAVs and similar slow-moving aerial targets, expanding the defensive perimeter and enabling earlier warnings. Full article

This paper proposes a YOLO fire detection algorithm based on an attention-enhanced ghost mode, mixed convolutional pyramids, and flame-centre detection (AEGG-FD). Specifically, the enhanced ghost bottleneck is stacked to reduce redundant feature mapping operations in the process for achieving lightweight reconfiguration of the backbone, while attention is added to compensate for accuracy loss. Furthermore, a feature pyramid built using mixed convolution is introduced to accelerate network inference speed. Finally, the local information is extracted by the designed flame-centre detection (FD) module for furnishing auxiliary information in effective firefighting. Experimental results on both the benchmark fire dataset and the video dataset show that the AEGG-FD performs better than the classical YOLO-based models such as YOLOv5, YOLOv7 and YOLOv8. Specifically, both the mean accuracy (mAP_0.5, reaching 84.7%) and the inferred speed (FPS) are improved by 6.5 and 8.4 respectively, and both the number of model parameters and model size are compressed to 72.4% and 44.6% those of YOLOv5, respectively. Therefore, AEGG-FD achieves an effective balance between model weight, detection speed, and accuracy in firefighting. Full article

The timely and effective detection of forest fires is crucial for environmental and socio-economic protection. Existing deep learning models struggle to balance accuracy and a lightweight design. We introduce SWVR, a new lightweight deep learning algorithm. Utilizing the Reparameterization Vision Transformer (RepViT) and Simple Parameter-Free Attention Module (SimAM), SWVR efficiently extracts fire-related features with reduced computational complexity. It features a bi-directional fusion network combining top-down and bottom-up approaches, incorporates lightweight Ghost Shuffle Convolution (GSConv), and uses the Wise Intersection over Union (WIoU) loss function. SWVR achieves 79.6% accuracy in detecting forest fires, which is a 5.9% improvement over the baseline, and operates at 42.7 frames per second. It also reduces the model parameters by 11.8% and the computational cost by 36.5%. Our results demonstrate SWVR’s effectiveness in achieving high accuracy with fewer computational resources, offering practical value for forest fire detection. Full article

Human action recognition is a computer vision task that identifies how a person or a group acts on a video sequence. Various methods that rely on deep-learning techniques, such as two- or three-dimensional convolutional neural networks (2D-CNNs, 3D-CNNs), recurrent neural networks (RNNs), and vision transformers (ViT), have been proposed to address this problem over the years. Motivated by the fact that most of the used CNNs in human action recognition present high complexity, and the necessity of implementations on mobile platforms that are characterized by restricted computational resources, in this article, we conduct an extensive evaluation protocol over the performance metrics of five lightweight architectures. In particular, we examine how these mobile-oriented CNNs (viz., ShuffleNet-v2, EfficientNet-b0, MobileNet-v3, and GhostNet) execute in spatial analysis compared to a recent tiny ViT, namely EVA-02-Ti, and a higher computational model, ResNet-50. Our models, previously trained on ImageNet and BU101, are measured for their classification accuracy on HMDB51, UCF101, and six classes of the NTU dataset. The average and max scores, as well as the voting approaches, are generated through three and fifteen RGB frames of each video, while two different rates for the dropout layers were assessed during the training. Last, a temporal analysis via multiple types of RNNs that employ features extracted by the trained networks is examined. Our results reveal that EfficientNet-b0 and EVA-02-Ti surpass the other mobile-CNNs, achieving comparable or superior performance to ResNet-50. Full article

To solve the problem of the poor performance of a flame detection algorithm in a complex forest background, such as poor detection performance, insensitivity to small targets, and excessive computational load, there is an urgent need for a lightweight, high-accuracy, real-time detection system. This paper introduces a lightweight object-detection algorithm called GS-YOLOv5s, which is based on the YOLOv5s baseline model and incorporates a multi-scale feature fusion knowledge distillation architecture. Firstly, the ghost shuffle convolution bottleneck is applied to obtain richer gradient information through branching. Secondly, the WIoU loss function is used to address the issues of GIoU related to model optimization, slow convergence, and inaccurate regression. Finally, a knowledge distillation algorithm based on feature fusion is employed to further improve its accuracy. Experimental results based on the dataset show that compared to the YOLOv5s baseline model, the proposed algorithm reduces the number of parameters and floating-point operations by approximately 26% and 36%, respectively. Moreover, it achieved a 3.1% improvement in mAP_0.5 compared to YOLOv5s. The experiments demonstrate that GS-YOLOv5s, based on multi-scale feature fusion, not only enhances detection accuracy but also meets the requirements of lightweight and real-time detection in forest fire detection, commendably improving the practicality of flame-detection algorithms. Full article

This paper discusses the application of deep learning technology in recognizing vehicle black smoke in road traffic monitoring videos. The use of massive surveillance video data imposes higher demands on the real-time performance of vehicle black smoke detection models. The YOLOv5s model, known for its excellent single-stage object detection performance, has a complex network structure. Therefore, this study proposes a lightweight real-time detection model for vehicle black smoke, named MGSNet, based on the YOLOv5s framework. The research involved collecting road traffic monitoring video data and creating a custom dataset for vehicle black smoke detection by applying data augmentation techniques such as changing image brightness and contrast. The experiment explored three different lightweight networks, namely ShuffleNetv2, MobileNetv3 and GhostNetv1, to reconstruct the CSPDarknet53 backbone feature extraction network of YOLOv5s. Comparative experimental results indicate that reconstructing the backbone network with MobileNetv3 achieved a better balance between detection accuracy and speed. The introduction of the squeeze excitation attention mechanism and inverted residual structure from MobileNetv3 effectively reduced the complexity of black smoke feature fusion. Simultaneously, a novel convolution module, GSConv, was introduced to enhance the expression capability of black smoke features in the neck network. The combination of depthwise separable convolution and standard convolution in the module further reduced the model’s parameter count. After the improvement, the parameter count of the model is compressed to 1/6 of the YOLOv5s model. The lightweight vehicle black smoke real-time detection network, MGSNet, achieved a detection speed of 44.6 frames per second on the test set, an increase of 18.9 frames per second compared with the YOLOv5s model. The [email protected] still exceeded 95%, meeting the application requirements for real-time and accurate detection of vehicle black smoke. Full article

Deep-learning-based methods for plant disease recognition pose challenges due to their high number of network parameters, extensive computational requirements, and overall complexity. To address this issue, we propose an improved residual-network-based multi-plant disease recognition method that combines the characteristics of plant diseases. Our approach introduces a lightweight technique called maximum grouping convolution to the ResNet18 model. We made three enhancements to adapt this method to the characteristics of plant diseases and ultimately reduced the convolution kernel requirements, resulting in the final model, Model_Lite. The experimental dataset comprises 20 types of plant diseases, including 13 selected from the publicly available Plant Village dataset and seven self-constructed images of apple leaves with complex backgrounds containing disease symptoms. The experimental results demonstrated that our improved network model, Model_Lite, contains only about 1/344th of the parameters and requires 1/35th of the computational effort compared to the original ResNet18 model, with a marginal decrease in the average accuracy of only 0.34%. Comparing Model_Lite with MobileNet, ShuffleNet, SqueezeNet, and GhostNet, our proposed Model_Lite model achieved a superior average recognition accuracy while maintaining a much smaller number of parameters and computational requirements than the above models. Thus, the Model_Lite model holds significant potential for widespread application in plant disease recognition and can serve as a valuable reference for future research on lightweight network model design. Full article

Wheat leaf diseases are considered to be the foremost threat to wheat yield. In the realm of crop disease detection, convolutional neural networks (CNNs) have emerged as important tools. The training strategy and the initial learning rate are key factors that impact the performance and training speed of the model in CNNs. This study employed six training strategies, including Adam, SGD, Adam + StepLR, SGD + StepLR, Warm-up + Cosine annealing + SGD, Warm-up + Cosine, and annealing + Adam, with three initial learning rates (0.05, 0.01, and 0.001). Using the wheat stripe rust, wheat powdery mildew, and healthy wheat datasets, five lightweight CNN models, namely MobileNetV3, ShuffleNetV2, GhostNet, MnasNet, and EfficientNetV2, were evaluated. The results showed that upon combining the SGD + StepLR with the initial learning rate of 0.001, the MnasNet obtained the highest recognition accuracy of 98.65%. The accuracy increased by 1.1% as compared to that obtained with the training strategy with a fixed learning rate, and the size of the parameters was only 19.09 M. The above results indicated that the MnasNet was appropriate for porting to the mobile terminal and efficient for automatically identifying wheat leaf diseases. Full article

Forest fires rank among the costliest and deadliest natural disasters globally. Identifying the smoke generated by forest fires is pivotal in facilitating the prompt suppression of developing fires. Nevertheless, succeeding techniques for detecting forest fire smoke encounter persistent issues, including a slow identification rate, suboptimal accuracy in detection, and challenges in distinguishing smoke originating from small sources. This study presents an enhanced YOLOv8 model customized to the context of unmanned aerial vehicle (UAV) images to address the challenges above and attain heightened precision in detection accuracy. Firstly, the research incorporates Wise-IoU (WIoU) v3 as a regression loss for bounding boxes, supplemented by a reasonable gradient allocation strategy that prioritizes samples of common quality. This strategic approach enhances the model’s capacity for precise localization. Secondly, the conventional convolutional process within the intermediate neck layer is substituted with the Ghost Shuffle Convolution mechanism. This strategic substitution reduces model parameters and expedites the convergence rate. Thirdly, recognizing the challenge of inadequately capturing salient features of forest fire smoke within intricate wooded settings, this study introduces the BiFormer attention mechanism. This mechanism strategically directs the model’s attention towards the feature intricacies of forest fire smoke, simultaneously suppressing the influence of irrelevant, non-target background information. The obtained experimental findings highlight the enhanced YOLOv8 model’s effectiveness in smoke detection, proving an average precision (AP) of 79.4%, signifying a notable 3.3% enhancement over the baseline. The model’s performance extends to average precision small (APS) and average precision large (APL), registering robust values of 71.3% and 92.6%, respectively. Full article

The farming industry is facing the major challenge of intensive and inefficient harvesting labors. Thus, an efficient and automated fruit harvesting system is required. In this study, three object classification models based on Yolov5m integrated with BoTNet, ShuffleNet, and GhostNet convolutional neural networks (CNNs), respectively, are proposed for the automatic detection of tomato fruit. The various models were trained using 1508 normalized images containing three classes of cherry tomatoes, namely ripe, immature, and damaged. The detection accuracy for the three classes was found to be 94%, 95%, and 96%, respectively, for the modified Yolov5m + BoTNet model. The model thus appeared to provide a promising basis for the further development of automated harvesting systems for tomato fruit. Full article

Forest fires pose significant hazards to ecological environments and economic society. The detection of forest fire smoke can provide crucial information for the suppression of early fires. Previous detection models based on deep learning have been limited in detecting small smoke and smoke with smoke-like interference. In this paper, we propose a lightweight model for forest fire smoke detection that is suitable for UAVs. Firstly, a smoke dataset is created from a combination of forest smoke photos obtained through web crawling and enhanced photos generated by using the method of synthesizing smoke. Secondly, the GSELAN and GSSPPFCSPC modules are built based on Ghost Shuffle Convolution (GSConv), which efficiently reduces the number of parameters in the model and accelerates its convergence speed. Next, to address the problem of indistinguishable feature boundaries between clouds and smoke, we integrate coordinate attention (CA) into the YOLO feature extraction network to strengthen the extraction of smoke features and attenuate the background information. Additionally, we use Content-Aware Reassembly of FEatures (CARAFE) upsampling to expand the receptive field in the feature fusion network and fully exploit the semantic information. Finally, we adopt SCYLLA-Intersection over Union (SIoU) loss as a replacement for the original loss function in the prediction phase. This substitution leads to improved convergence efficiency and faster convergence. The experimental results demonstrate that the LMDFS model proposed for smoke detection achieves an accuracy of 80.2% with a 5.9% improvement compared to the baseline and a high number of Frames Per Second (FPS)—63.4. The model also reduces the parameter count by 14% and Giga FLoating-point Operations Per second (GFLOPs) by 6%. These results suggest that the proposed model can achieve a high accuracy while requiring fewer computational resources, making it a promising approach for practical deployment in applications for detecting smoke. Full article

Maritime search and rescue is a crucial component of the national emergency response system, which mainly relies on unmanned aerial vehicles (UAVs) to detect objects. Most traditional object detection methods focus on boosting the detection accuracy while neglecting the detection speed of the heavy model. However, improving the detection speed is essential, which can provide timely maritime search and rescue. To address the issues, we propose a lightweight object detector named Shuffle-GhostNet-based detector (SG-Det). First, we construct a lightweight backbone named Shuffle-GhostNet, which enhances the information flow between channel groups by redesigning the correlation group convolution and introducing the channel shuffle operation. Second, we propose an improved feature pyramid model, namely BiFPN-tiny, which has a lighter structure capable of reinforcing small object features. Furthermore, we incorporate the Atrous Spatial Pyramid Pooling module (ASPP) into the network, which employs atrous convolution with different sampling rates to obtain multi-scale information. Finally, we generate three sets of bounding boxes at different scales—large, medium, and small—to detect objects of different sizes. Compared with other lightweight detectors, SG-Det achieves better tradeoffs across performance metrics and enables real-time detection with an accuracy rate of over 90% for maritime objects, showing that it can better meet the actual requirements of maritime search and rescue. Full article

The detection of coal and gangue is an essential part of intelligent sorting. A lightweight coal and gangue detection algorithm based on You Only Look Once version 5s (Yolov5s) is proposed for the current coal and gangue target detection algorithm with the low accuracy of small target detection, high model complexity, and sizeable computational memory consumption. Firstly, we build a new convolutional block based on the Funnel Rectified Linear Unit (FReLU) activation function and apply it to the original Yolov5s network so that the model adaptively captures local contextual information of the image. Secondly, the neck of the original network is redesigned to improve the detection accuracy of small samples by adding a small target detection head to achieve multi-scale feature fusion. Next, some of the standard convolution modules in the original network are replaced with Depthwise Convolution (DWC) and Ghost Shuffle Convolution (GSC) modules to build a lightweight feature extraction network while ensuring the model detection accuracy. Finally, an efficient channel attention (ECA) module is embedded in the backbone of the lightweight network to facilitate accurate localization of the prediction region by improving the information interaction of the model with the channel features. In addition, the importance of each component is fully demonstrated by ablation experiments and visualization analysis comparison experiments. The experimental results show that the mean average precision (mAP) and the model size of our proposed model reach 0.985 and 4.9 M, respectively. The mAP is improved by 0.6%, and the number of parameters is reduced by 72.76% compared with the original Yolov5s network. The improved algorithm has higher localization and recognition accuracy while significantly reducing the number of floating-point calculations and of parameters, reducing the dependence on hardware, and providing a specific reference basis for deploying automated underground gangue sorting. Full article

Winter jujube is a popular fresh fruit in China for its high vitamin C nutritional value and delicious taste. In terms of winter jujube object detection, in machine learning research, small size jujube fruits could not be detected with a high accuracy. Moreover, in deep learning research, due to the large model size of the network and slow detection speed, deployment in embedded devices is limited. In this study, an improved Yolov5s (You Only Look Once version 5 small model) algorithm was proposed in order to achieve quick and precise detection. In the improved Yolov5s algorithm, we decreased the model size and network parameters by reducing the backbone network size of Yolov5s to improve the detection speed. Yolov5s’s neck was replaced with slim-neck, which uses Ghost-Shuffle Convolution (GSConv) and one-time aggregation cross stage partial network module (VoV-GSCSP) to lessen computational and network complexity while maintaining adequate accuracy. Finally, knowledge distillation was used to optimize the improved Yolov5s model to increase generalization and boost overall performance. Experimental results showed that the accuracy of the optimized Yolov5s model outperformed Yolov5s in terms of occlusion and small target fruit discrimination, as well as overall performance. Compared to Yolov5s, the Precision, Recall, mAP (mean average Precision), and F₁ values of the optimized Yolov5s model were increased by 4.70%, 1.30%, 1.90%, and 2.90%, respectively. The Model size and Parameters were both reduced significantly by 86.09% and 88.77%, respectively. The experiment results prove that the model that was optimized from Yolov5s can provide a real time and high accuracy small winter jujube fruit detection method for robot harvesting. Full article

The detection of apple yield in complex orchards plays an important role in smart agriculture. Due to the large number of fruit trees in the orchard, improving the speed of apple detection has become one of the challenges of apple yield detection. Additional challenges in the detection of apples in complex orchard environments are vision obstruction by leaves, branches and other fruit, and uneven illumination. The YOLOv5 (You Only Look Once version 5) network structure has thus far been increasingly utilized for fruit recognition, but its detection accuracy and real-time detection speed can be improved. Thus, an upgraded lightweight apple detection method YOLOv5-PRE (YOLOv5 Prediction) is proposed for the rapid detection of apple yield in an orchard environment. The ShuffleNet and the GhostNet lightweight structures were introduced into the YOLOv5-PRE model to reduce the size of the model, and the CA (Coordinate Attention) and CBAM (Convolutional Block Attention Module) attention mechanisms were used to improve the detection accuracy of the algorithm. After applying this algorithm on PC with NVIDIA Quadro P620 GPU, and after comparing the results of the YOLOv5s (You Only Look Once version 5 small) and the YOLOv5-PRE models outputs, the following conclusions were obtained: the average precision of the YOLOv5-PRE model was 94.03%, which is 0.58% higher than YOLOv5s. As for the average detection time of a single image on GPU and CPU, it was 27.0 ms and 172.3 ms, respectively, which is 17.93% and 35.23% higher than YOLOV5s. Added to that, the YOLOv5-PRE model had a missed detection rate of 6.54% when being subject to back-light conditions, and a false detection rate of 4.31% when facing front-light conditions, which are 2.8% and 0.86% higher than YOLOv5s, respectively. Finally, the feature extraction process of the YOLOv5-PRE model was presented in the form of a feature map visualization, which enhances the interpretability of the model. Thus, the YOLOv5-PRE model is more suitable for transplanting into embedded devices and adapts well to different lighting conditions in the orchard, which provides an effective method and a theoretical basis for the rapid detection of apples in the process of rapid detection of apple yield. Full article