Search

Wildfires pose significant environmental and societal threats, necessitating improved early detection methods. This study investigates the effectiveness of integrating real-time object detection deep learning models (YOLOv8 and RT-DETR) with advanced data augmentation techniques, including StyleGAN2-ADA, for wildfire smoke detection. We evaluated model performance on datasets enhanced with fundamental transformations and synthetic images, focusing on detection accuracy. YOLOv8X demonstrated superior overall performance with AP@0.33 of 0.962 and AP@0.5 of 0.900, while RT-DETR-X excelled in small object detection with a 0.983 detection rate. Data augmentation, particularly StyleGAN2-ADA, significantly enhanced model performance across various metrics. Our approach reduced average detection times to 1.52 min for YOLOv8X and 2.40 min for RT-DETR-X, outperforming previous methods. The models demonstrated robust performance under challenging conditions, like fog and camera noise, providing reassurance of their effectiveness. While false positives remain a challenge, these advancements contribute significantly to early wildfire smoke detection capabilities, potentially mitigating wildfire impacts through faster response times. This research establishes a foundation for more effective wildfire management strategies and underscores the potential of deep learning applications in environmental monitoring.

Convolutional Neural Networks (CNNs) are well-studied and commonly used for the problem of object detection thanks to their increased accuracy. However, high accuracy on its own says little about the effective performance of CNN-based models, especially when real-time detection tasks are involved. To the best of our knowledge, there has not been sufficient evaluation of the available methods in terms of their speed/accuracy trade-off. This work performs a review and hands-on evaluation of the most fundamental object detection models on the Common Objects in Context (COCO) dataset with respect to this trade-off, their memory footprint, and computational and storage costs. In addition, we review available datasets for medical mask detection and train YOLOv5 on the Properly Wearing Masked Faces Dataset (PWMFD). Next, we test and evaluate a set of specific optimization techniques, transfer learning, data augmentations, and attention mechanisms, and we report on their effect for real-time mask detection. Based on our findings, we propose an optimized model based on YOLOv5s using transfer learning for the detection of correctly and incorrectly worn medical masks that surpassed more than two times in speed (69 frames per second) the state-of-the-art model SE-YOLOv3 on the PWMFD while maintaining the same level of mean Average Precision (67%).

Deep learning models have significantly improved object detection, which is essential for visual sensing. However, their increasing complexity results in higher latency and resource consumption, making real-time object detection challenging. In order to address the challenge, we propose a new lightweight filtering method called L-filter to predict empty video frames that include no object of interest (e.g., vehicles) with high accuracy via hybrid time series analysis. L-filter drops those frames deemed empty and conducts object detection for nonempty frames only, significantly enhancing the frame processing rate and scalability of real-time object detection. Our evaluation demonstrates that L-filter improves the frame processing rate by 31–47% for a single traffic video stream compared to three standalone state-of-the-art object detection models without L-filter. Additionally, L-filter significantly enhances scalability; it can process up to six concurrent video streams in one commodity GPU, supporting over 57 fps per stream, by working alongside the fastest object detection model among the three models.

Deep learning technology for real-time small object detection in aerial images can be used in various industrial environments such as real-time traffic surveillance and military reconnaissance. However, detecting small objects with few pixels and low resolution remains a challenging problem that requires performance improvement. To improve the performance of small object detection, we propose DCEF 2-YOLO. Our proposed method enables efficient real-time small object detection by using a deformable convolution (DFConv) module and an efficient feature fusion structure to maximize the use of the internal feature information of objects. DFConv preserves small object information by preventing the mixing of object information with the background. The optimized feature fusion structure produces high-quality feature maps for efficient real-time small object detection while maximizing the use of limited information. Additionally, modifying the input data processing stage and reducing the detection layer to suit small object detection also contributes to performance improvement. When compared to the performance of the latest YOLO-based models (such as DCN-YOLO and YOLOv7), DCEF 2-YOLO outperforms them, with a mAP of +6.1% on the DOTA-v1.0 test set, +0.3% on the NWPU VHR-10 test set, and +1.5% on the VEDAI512 test set. Furthermore, it has a fast processing speed of 120.48 FPS with an RTX3090 for 512 × 512 images, making it suitable for real-time small object detection tasks.

In the field of unmanned systems, the combination of artificial intelligence with self-operating functionalities is becoming increasingly important. This study introduces a new method for autonomously detecting humans in indoor environments using unmanned aerial vehicles, utilizing the advanced techniques of a deep learning framework commonly known as “You Only Look Once” (YOLO). The key contribution of this research is the development of a new model (YOLO-IHD), specifically designed for human detection in indoor using drones. This model is created using a unique dataset gathered from aerial vehicle footage in various indoor environments. It significantly improves the accuracy of detecting people in these complex environments. The model achieves a notable advancement in autonomous monitoring and search-and-rescue operations, highlighting its importance for tasks that require precise human detection. The improved performance of the new model is due to its optimized convolutional layers and an attention mechanism that process complex visual data from indoor environments. This results in more dependable operation in critical situations like disaster response and indoor rescue missions. Moreover, when combined with an accelerating processing library, the model shows enhanced real-time detection capabilities and operates effectively in a real-world environment with a custom designed indoor drone. This research lays the groundwork for future enhancements designed to significantly increase the model’s accuracy and the reliability of indoor human detection in real-time drone applications.

Home service robots operating indoors, such as inside houses and offices, require the real-time and accurate identification and location of target objects to perform service tasks efficiently. However, images captured by visual sensors while in motion states usually contain varying degrees of blurriness, presenting a significant challenge for object detection. In particular, daily life scenes contain small objects like fruits and tableware, which are often occluded, further complicating object recognition and positioning. A dynamic and real-time object detection algorithm is proposed for home service robots. This is composed of an image deblurring algorithm and an object detection algorithm. To improve the clarity of motion-blurred images, the DA-Multi-DCGAN algorithm is proposed. It comprises an embedded dynamic adjustment mechanism and a multimodal multiscale fusion structure based on robot motion and surrounding environmental information, enabling the deblurring processing of images that are captured under different motion states. Compared with DeblurGAN, DA-Multi-DCGAN had a 5.07 improvement in Peak Signal-to-Noise Ratio (PSNR) and a 0.022 improvement in Structural Similarity (SSIM). An AT-LI-YOLO method is proposed for small and occluded object detection. Based on depthwise separable convolution, this method highlights key areas and integrates salient features by embedding the attention module in the AT-Resblock to improve the sensitivity and detection precision of small objects and partially occluded objects. It also employs a lightweight network unit Lightblock to reduce the network’s parameters and computational complexity, which improves its computational efficiency. Compared with YOLOv3, the mean average precision (mAP) of AT-LI-YOLO increased by 3.19%, and the detection precision of small objects, such as apples and oranges and partially occluded objects, increased by 19.12% and 29.52%, respectively. Moreover, the model inference efficiency had a 7 ms reduction in processing time. Based on the typical home activities of older people and children, the dataset Grasp-17 was established for the training and testing of the proposed method. Using the TensorRT neural network inference engine of the developed service robot prototype, the proposed dynamic and real-time object detection algorithm required 29 ms, which meets the real-time requirement of smooth vision.

The accuracy of current deep learning algorithms has certainly increased. However, deploying deep learning networks on edge devices with limited resources is challenging due to their inherent depth and high parameter count. Here, we proposed an improved YOLO model based on an attention mechanism and receptive field (RFA-YOLO) model, applying the MobileNeXt network as the backbone to reduce parameters and complexity, adopting the Receptive Field Block (RFB) and Efficient Channel Attention (ECA) modules to improve the detection accuracy of multi-scale and small objects. Meanwhile, an FPGA-based model deployment solution was proposed to implement parallel acceleration and low-power deployment of the detection algorithm model, which achieved real-time object detection for optical remote sensing images. We implement the proposed DPU and Vitis AI-based object detection algorithms with FPGA deployment to achieve low power consumption and real-time performance requirements. Experimental results on DIOR dataset demonstrate the effectiveness and superiority of our RFA-YOLO model for object detection algorithms. Moreover, to evaluate the performance of the proposed hardware implementation, it was implemented on a Xilinx ZCU104 board. Results of the experiments for hardware and software simulation show that our DPU-based hardware implementation are more power efficient than central processing units (CPUs) and graphics processing units (GPUs), and have the potential to be applied to onboard processing systems with limited resources and power consumption.

In this paper, a deep learning enabled object detection model for multi-class plant disease has been proposed based on a state-of-the-art computer vision algorithm. While most existing models are limited to disease detection on a large scale, the current model addresses the accurate detection of fine-grained, multi-scale early disease detection. The proposed model has been improved to optimize for both detection speed and accuracy and applied to multi-class apple plant disease detection in the real environment. The mean average precision (mAP) and F1-score of the detection model reached up to 91.2% and 95.9%, respectively, at a detection rate of 56.9 FPS. The overall detection result demonstrates that the current algorithm significantly outperforms the state-of-the-art detection model with a 9.05% increase in precision and 7.6% increase in F1-score. The proposed model can be employed as an effective and efficient method to detect different apple plant diseases under complex orchard scenarios.

One of the tasks of robotics is to develop a robot’s ability to perform specific actions for as long as possible without human assistance. One such step is to open different types of doors. This task is essential for any operation that involves moving a robot from one room to another. This paper proposes a versatile and computationally efficient algorithm for an autonomous mobile robot opening different types of doors, using machine learning methods. The latter include the YOLOv5 object detection model, the RANSAC iterative method for estimating the mathematical model parameters, and the DBSCAN clustering algorithm. Alternative clustering methods are also compared. The proposed algorithm was explored and tested in simulation and on a real robot manufactured by SOMATIC version Dalek. The percentage of successful doors opened out of the total number of attempts was used as an accuracy metric. The proposed algorithm reached an accuracy of 95% in 100 attempts. The result of testing the door-handle detection algorithm on simulated data was an error of 1.98 mm in 10,000 samples. That is, the average distance from the door handle found by the detector to the real one was 1.98 mm. The proposed algorithm has shown high accuracy and the ability to be applied in real time for opening different types of doors.

Transformers have become increasingly prevalent in computer vision research, especially for object detection. To accurately and efficiently distinguish the stem end of pomelo from its black spots, we propose a hierarchical feature detector, which reconfigures the self-attention model, with high detection accuracy. We designed the combination attention module and the hierarchical feature fusion module that utilize multi-scale features to improve detection performance. We created a dataset in COCO format and annotated two types of detection targets: the stem end and the black spot. Experimental results on our pomelo dataset confirm that HFD’s results are comparable to those of state-of-the-art one-stage detectors such as YOLO v4 and YOLO v5 and transformer-based detectors such as DETR, Deformable DETR, and YOLOS. It achieves 89.65% mAP at 70.92 FPS with 100.34 M parameters.

Object detection is one of the most popular areas today. The new models of object detection are created continuously and applied in various fields that help to modernize the old solutions in practice. In this manuscript, the focus has been on investigating the influence of training parameters on similar object detection: image resolution, batch size, iteration number, and color of images. The results of the model have been applied in real-time object detection using mobile devices. The new construction detail dataset has been collected and used in experimental investigation. The models have been evaluated by two measures: the accuracy of each prepared model has been measured; results of real-time object detection on testing data, where the recognition ratio has been calculated. The highest influence on the accuracy of the created models has the iteration number chosen in the training process and the resolution of the images. The higher the resolution of the images that have been selected, the lower the accuracy that has been obtained. The small iteration number leads to the model not being well trained and the accuracy of the models being very low. Slightly better results were obtained when the color images were used.

At present, the surveillance of river floating in China is labor-intensive, time-consuming, and may miss something, so a fast and accurate automatic detection method is necessary. The two-stage convolutional neural network models appear to have high detection accuracy, but it is hard to reach real-time detection, while on the other hand, the one-stage models are less time-consuming but have lower accuracy. In response to the above problems, we propose a one-stage object detection model EYOLOv3 to achieve real-time and high accuracy detection of floating objects in video streams. Firstly, we design a multi-scale feature extraction and fusion module to improve the feature extraction capability of the network. Secondly, a better clustering algorithm is used to analyze the size characteristics of floating objects to design the anchor box, enabling the network to detect objects more effectively. Then a focus loss function is proposed to make the network effectively overcome the sample imbalance problem, and finally, an improved NMS algorithm is proposed to solve the object suppressed problem. Experiments show that the proposed model is efficient in detection of river floating objects, and has better performance than the classical object detection method and the latest method, realizing real-time floating detection in video streams.

Wheat head detection can estimate various wheat traits, such as density, health, and the presence of wheat head. However, traditional detection methods have a huge array of problems, including low efficiency, strong subjectivity, and poor accuracy. In this paper, a method of wheat-head detection based on a deep neural network is proposed to enhance the speed and accuracy of detection. The YOLOv4 is taken as the basic network. The backbone part in the basic network is enhanced by adding dual spatial pyramid pooling (SPP) networks to improve the ability of feature learning and increase the receptive field of the convolutional network. Multilevel features are obtained by a multipath neck part using a top-down to bottom-up strategy. Finally, YOLOv3′s head structures are used to predict the boxes of wheat heads. For training images, some data augmentation technologies are used. The experimental results demonstrate that the proposed method has a significant advantage in accuracy and speed. The mean average precision of our method is 94.5%, and the detection speed is 71 FPS that can achieve the effect of real-time detection.

The USV (unmanned surface vehicle) is playing an important role in many tasks such as marine environmental observation and maritime security, for the advantages of high autonomy and mobility. Detecting the targets on the surface of the water with high precision ensures the subsequent task implementation. However, the changes from the lights and the surface environment influence the performance of the target detecting method in a long-term task with USV. Therefore, this paper proposed a novel target detection method by fusing DenseNet in YOLOV3 to improve the stability of detection to decrease the feature loss, while the target feature is transmitted in the layers of a deep neural network. All the image data used to train and test the proposed method were obtained in the real ocean environment with a USV in the South China Sea during a one month sea trial in November 2019. The experiment results demonstrate the performance of the proposed method is more suitable for the changed weather conditions though comparing with the existing methods, and the real-time performance is available in practical ocean tasks for USV.

As unmanned ground and aerial vehicles become more accessible and their usage covers a wider area of application, including for threatening purposes which can cause connected catastrophe, a surveillance system for the public places is being considered more essential to respond to those possible threats. We propose an inexpensive, lighter, safer, and smaller radar system than military-grade radar systems while keeping reasonable capability for use in monitoring public places. The paper details the iterative process on the system design and improvements with experiments to realize the system used for surveillance. The experiments show the practical use of the system and configuration for a better understanding of using the system. Cyber-physical systems for outdoor environments can benefit from the system as a sensor for sensing objects as well as monitoring.

The Single Shot MultiBox Detector (SSD) is one of the fastest algorithms in the current target detection field. It has achieved good results in target detection but there are problems such as poor extraction of features in shallow layers and loss of features in deep layers. In this paper, we propose an accurate and efficient target detection method, named Single Shot Object Detection with Feature Enhancement and Fusion (FFESSD), which is to enhance and exploit the shallow and deep features in the feature pyramid structure of the SSD algorithm. To achieve it we introduced the Feature Fusion Module and two Feature Enhancement Modules, and integrated them into the conventional structure of the SSD. Experimental results on the PASCAL VOC 2007 dataset demonstrated that FFESSD achieved 79.1% mean average precision (mAP) at the speed of 54.3 frame per second (FPS) with the input size 300 × 300, while FFESSD with a 512 × 512 sized input achieved 81.8% mAP at 30.2 FPS. The proposed network shows state-of-the-art mAP, which is better than the conventional SSD, Deconvolutional Single Shot Detector (DSSD), Feature-Fusion SSD (FSSD), and other advanced detectors. On extended experiment, the performance of FFESSD in fuzzy target detection was better than the conventional SSD.