Real-Time Water Surface Object Detection Based on Improved Faster R-CNN.

With the rise of Deep Learning (DL), numerous DL-based SAR ship detectors, represented by Faster R-CNN, is constantly breaking the record of detection accuracy. However, these detectors still face huge challenges in modeling the geometric transformation of shape-changeable ships, due to their used conventional convolution kernels whose structure is fixed. Therefore, to address this problem, we propose an improved Faster R-CNN by using deformable convolution kernels for SAR ship detection. We substitute some conventional shape-changeless convolution kernels in Faster R-CNN with deformable convolution ones that can adaptively learn additional 2-D offsets of the raw convolution kernels, to better model the geometric transformation of shape-changeable ships. Finally, the experimental results on the open SAR Ship Detection Dataset (SSDD) reveal that our improved Faster R-CNN achieves a 2.02% mean Average Precision (mAP) improvement than the raw Faster R-CNN.

Recent research on neocognitron-like neural feed-forward architectures, which have formerly been successfully applied to the recognition of artificial stimuli such as paperclip objects, now also opens up application to more natural stimuli. Such networks exhibit high-recognition performance with respect to translation, rotation, scaling and cluttered surroundings. In this contribution, we introduce a new type of hierarchical model, which is trained using a non-negative matrix factorization algorithm. In contrast to previous work, our approach cannot only classify objects but is also capable of rapid object detection in natural scenes. Thus, the time-consuming and conceptually unsatisfying split-up into a localization stage (e.g., using segmentation) and a subsequent classification can be avoided. The network consists of alternating layers of simple and complex cell planes and incorporates nonlinear processing schemes that have been proposed in recent literature. Learning of receptive field profiles for the lower layers of the network takes place by unsupervised learning whereas a final classification layer is trained supervised. This final layer is then utilized for detection. We test the classification performance of the network on images of natural objects which are systematically distorted. To test the ability to detect objects, cluttered natural background is used.

We propose a novel edge detection method based on fractal coding. We point out that the mean-square-error distance (MSED) of fractal coding can be used to extract edges of a fractal image effectively. The algorithm for edge detection based on fractal coding MSED is given and its improvement is discussed. Results indicate that since the new algorithm for edge detection avoids the gradient operation, it is insensitive to noise and suitable for object detection in natural scenes based on edges.

Multi-scale object detection in natural scenes is still challenging. To enhance the multi-scale perception capability, some algorithms combine the lower-level and higher-level information via multi-scale feature fusion strategies. However, the inherent spatial properties among instances and relations between foreground and background are ignored. In addition, the human-defined “center-based” regression quality evaluation strategy, predicting a high-to-low score based on a linear relationship with the distance to the center of ground-truth box, is not robust to scale-variant objects. In this work, we propose a Depth-Guided Progressive Network (DGPNet) for multi-scale object detection. Specifically, besides the prediction of classification and localization, the depth is estimated and used to guide the image features in a weighted manner to obtain a better spatial representation. Therefore, depth estimation and 2D object detection are simultaneously learned via a unified network, where the depth features are merged as auxiliary information into the detection branch to enhance the discrimination among multi-scale objects. Moreover, to overcome the difficulty of empirically fitting the localization quality function, high-quality predicted boxes on scale-variant objects are more adaptively obtained by an IoU-aware progressive sampling strategy. We divide the sampling process into two stages, i.e., “statistical-aware” and “IoU-aware”. The former selects thresholds for positive samples based on statistical characteristics of multi-scale instances, and the latter further selects high-quality samples by IoU on the basis of the former. Therefore, the final ranking scores better reflect the quality of localization. Experiments verify that our method outperforms state-of-the-art methods on the KINS and Cityscapes dataset.

To enhance the working efficiency of apple picking systems by improving the accuracy of object detection in natural scenes, an improved apple detection network based on YOLOX-s is proposed. The self-attention residual module is added to the last layer of the improved YOLOX-s, which is used to add global feature information for small apples. An additional object detection head is added in YOLOX-s to strengthen the feature information of dense objects. Extra convolutional branches are added into the spatial pyramid pooling structure to enhance feature fusion and compensate for lost object location information. The loss function of the network is replaced by the α-CIoU loss, which is used to improve the bbox regression accuracy by up-weighting the loss and gradient of the high IoU prediction box. Experiment results show that the mAP50 value and the recall rate of the improved network reached 91.8% and 97%, respectively. Therefore, the improved network is superior to the existing detection networks in detection accuracy. Meanwhile, the detection time increased slightly, but it still meets the requirements of real-time detection of the picking system.

Apple leaf diseases seriously affect the sustainable production of apple fruit. Early infection monitoring of apple leaves and timely disease control measures are the key to ensuring the regular growth of apple fruits and achieving a high-efficiency economy. Consequently, disease detection schemes based on computer vision can compensate for the shortcomings of traditional disease detection methods that are inaccurate and time-consuming. Nowadays, to solve the limitations ranging from complex background environments to dense and small characteristics of apple leaf diseases, an improved Faster region-based convolutional neural network (Faster R-CNN) method was proposed. The advanced Res2Net and feature pyramid network architecture were introduced as the feature extraction network for extracting reliable and multi-dimensional features. Furthermore, RoIAlign was also employed to replace RoIPool so that accurate candidate regions will be produced to address the object location. Moreover, soft non-maximum suppression was applied for precise detection performance of apple leaf disease when making inferences to the images. The improved Faster R-CNN structure behaves effectively in the annotated apple leaf disease dataset with an accuracy of 63.1% average precision, which is higher than other object detection methods. The experiments proved that our improved Faster R-CNN method provides a highly precise apple leaf disease recognition method that could be used in real agricultural practice.

With the increasing popularity of practical vision systems and smart phones, text detection in natural scenes becomes a critical yet challenging task. Most existing methods have focused on detecting horizontal or near-horizontal texts. In this paper, we propose a system which detects texts of arbitrary orientations in natural images. Our algorithm is equipped with a two-level classification scheme and two sets of features specially designed for capturing both the intrinsic characteristics of texts. To better evaluate our algorithm and compare it with other competing algorithms, we generate a new dataset, which includes various texts in diverse real-world scenarios; we also propose a protocol for performance evaluation. Experiments on benchmark datasets and the proposed dataset demonstrate that our algorithm compares favorably with the state-of-the-art algorithms when handling horizontal texts and achieves significantly enhanced performance on texts of arbitrary orientations in complex natural scenes.

Brace sleeve screws are an important component of catenary support devices in high-speed railways. According to the statistical data, there is a much risk for failure of brace sleeve screws during the operation of the catenary system. In order to obtain better results on the fault diagnosis for brace sleeve screws, it is of vital importance to develop a high-precision detection approach for brace sleeve screws. With the aim of solving the problems that the size of the brace sleeve screw is too small and the employed previous methods are difficult to detect accurately, a novel object detection network based on an improved Faster RCNN is proposed. Through the analysis of the factors that influence the detection accuracy for brace sleeve screws and the characteristics of original Faster RCNN, the concepts of discrimination maps and proposal maps are introduced for the improved Faster RCNN. In order to retain the characteristic information of brace sleeve screws, the high-resolution discrimination maps are adopted to identify the object type and the low-resolution proposal maps are adopted to generate the candidate boxes of region with an appropriate anchor scale configuration. Experimental results show that the proposed method has a higher detection accuracy for brace sleeve screws. In addition, compared with original Faster RCNN implementation based on the VGG-16 and ResNet-101 backbone, the detection accuracy of the improved Faster RCNN is increased by 8%. It also provides a new idea for general object detection of small targets.

We propose a lightweight detection algorithm based on the Single Shot MultiBox Detector (SSD) algorithm in order to facilitate sheep management and to realize sheep facial identification, and we take the self-constructed dataset as the research object. First, the SSD replaces the VGG16 backbone network with MobileNetv3, a lightweight neural network, to create a hybrid model that is much smaller. Second, the ECA attention mechanism is incorporated into the backend of the 72 × 160 bottleneck layer. Finally, the SmoothL1 loss function is substituted with the BalancedL1 loss function. The optimized model’s size decreases significantly from the original SSD’s 132 MB to just 22.4 MB. It achieves a mean average precision of 83.47% and maintains an average frame rate of 68.53 frames per second. Compared to the basic SSD model, the mean average precision has increased by 3.25 percentage points, the model size has decreased by 109.6 MB, and the detection speed has improved by 9.55 frames per second. In comparative experiments using the same dataset with different object detection models, the proposed model outperforms the SSD, Faster R-CNN, Retinanet, and CenterNet in terms of mean average precision, with improvements of 3.25 percentage points, 4.71 percentage points, 2.38 percentage points, and 8.13 percentage points, respectively. The detection speed has shown significant improvements, increasing by 9.55, 58.55, 53.1, and 12.37 frames per second, respectively. The improved model presented in this paper significantly reduces the model’s size and computational requirements while maintaining an excellent performance. This provides a valuable reference for the digitalization of animal husbandry and livestock farming.

Over the past few years, deep learning has played a tremendous role in real time object detection, recognition and classification. Today, algorithms use deep learning not only for simple object detection but have also paved way for vehicle detection that is complex situation such as Indian roads. Different deep learning models can learn features of varied complexity and perform according to surrounding environment in which they are deployed. In this paper a comparison of six models is performed for detection of vehicles on roads that are unstructured or unconstrained. Till date, deep learning algorithms are being used for vehicle (car, bus, truck, motorcycle etc.) detection. A dataset based on Indian road scenario has been used for training on models: faster RCNN and Single Shot Detector (SSD) combined with convolution classifiers mobilenet, resnet and inceptionnet. The six models are compared by evaluating mean average precision (mAP) and classification time complexity for IDD dataset. The observations suggested faster RCNN Resnet50 resolves the trade-off of time and accuracy. Whereas SSD mobilenet V1 takes 31.89% less time than faster RCNN resnet101. On the flipped side, faster RCNN resnet101is 34.38% more accurate than SSD mobilenet v1.

In order to ensure the safety of workers and the stable operation of the power grid, the power grid companies in China have developed a very strict safety control system which contains many regulations, such as safety regulations and the two-ticket regulations. However, some workers are still lack of safety awareness in that they even do not wear safety helmets when carrying out construction or maintenance projects in substations. Safety helmet is an indispensable safety tool in electric power work, which can maintain the head safety of workers at all times and avoid fatal injuries such as electric shock and strike. Working without safety helmet is not only a violation of the safety control system, but also a manifestation of not being responsible for personal life and property. Nevertheless, the existing control means can not identify and prevent such behavior timely, efficiently and accurately. In order to better avoid this unsafe behavior, this paper proposes the Improved Faster R-CNN algorithm to inspect the wearing of safety helmet. Considering the real situation, the Retinex image enhancement is introduced to improve image quality for the outdoor complex scenes in substations. K-means++ algorithm is also adopted for better adaptation to the small size helmet. The experimental results show that compared with the Faster R-CNN algorithm, the mean average precision of the Improved Faster R-CNN is improved and the real-time automatic detection of the wearing of safety helmets is realized.

The real-time detection of objects and persons in their approach remains vital for applications that need safety measures security protocols and efficiency in autonomous systems and surveillance operations and industrial automation. The proposed research creates a durable real-time detection system through sensor combination using ultrasonic along with infrared and LiDAR and radar sensors and computer vision-based sensors to enhance detection accuracy and response times in changing environmental scenarios. The method starts by processing data gathered from arrays of multiple sensors through optimized filtering techniques and extraction methods before machines learn to classify objects while reducing false detection instances. The assessment of performance takes place in different environmental conditions to track both detection speed along with accuracy levels and system reliability. Sensor fusion enables real-time detection systems to perform better because it reduces the impact of environmental changes and sensor distortions. System capabilities increase in different operational areas when researchers demonstrate pre-processing optimization works alongside intelligent data processing systems. The research findings combine autonomous navigation systems with smart surveillance technology together with safety protocols for industrial security which enable real-time scalability and reduced risk elements. Research demonstrates a specific achievement through its analysis that machines learn better when combined with multi-sensor integration over single-sensor models which improves system reliability when used in real-world applications. Multi-sensor arrays and advanced processing techniques generate enhanced system responsiveness alongside increased adaptability and reliability thus improving detection systems within industrial security measures and automation operations.

To solve the problems of high labor intensity, low efficiency, and frequent errors in the manual identification of cone yarn types, in this study five kinds of cone yarn were taken as the research objects, and an identification method for cone yarn based on the improved Faster R-CNN model was proposed. In total, 2750 images were collected of cone yarn samples in real of textile industry environments, then data enhancement was performed after marking the targets. The ResNet50 model with strong representation ability was used as the feature network to replace the VGG16 backbone network in the original Faster R-CNN model to extract the features of the cone yarn dataset. Training was performed with a stochastic gradient descent approach to obtain an optimally weighted file to predict the categories of cone yarn. Using the same training samples and environmental settings, we compared the method proposed in this paper with two mainstream target detection algorithms, YOLOv3 + DarkNet-53 and Faster R-CNN + VGG16. The results showed that the Faster R-CNN + ResNet50 algorithm had the highest mean average precision rate for the five types of cone yarn at 99.95%, as compared with the YOLOv3 + DarkNet-53 algorithm with a mean average precision rate that was 2.24% higher and the Faster R-CNN + VGG16 algorithm with a mean average precision that was 1.19% higher. Regarding cone yarn defects, shielding, and wear, the Faster R-CNN + ResNet50 algorithm can correctly identify these issues without misdetection occurring, with an average precision rate greater than 99.91%.

Real-time object detection in indoor settings is a challenging area of computer vision, faced with unique obstacles such as variable lighting and complex backgrounds. This field holds significant potential to revolutionize applications like augmented and mixed realities by enabling more seamless interactions between digital content and the physical world. However, the scarcity of research specifically fitted to the intricacies of indoor environments has highlighted a clear gap in the literature. To address this, our study delves into the evaluation of existing datasets and computational models, leading to the creation of a refined dataset. This new dataset is derived from OpenImages v7[14], focusing exclusively on 32 indoor categories selected for their relevance to real-world applications. Alongside this, we present an adaptation of a CNN detection model, incorporating an attention mechanism to enhance the model’s ability to discern and prioritize critical features within cluttered indoor scenes. Our findings demonstrate that this approach is not just competitive to existing state-of-the-art models in accuracy and speed but also opens new avenues for research and application in the field of real-time indoor object detection.

With the rapid development of autonomous driving, real-time object detection on 360° images becomes more and more important. In this paper, we propose a panoramic virtual dataset for training object detectors on 360° images. The most important feature of our dataset includes (1) an auto-generated city scene is created for rendering 360° dataset. (2) annotation work for this dataset is automatic. In addition, we propose a modified YOLOv3 model called Pano-YOLO for real-time panoramic object detection. Compared with YOLOv3, mAP of Pano-YOLO drops 0.39%. While speed is 32.47% faster. Experiments are performed to show that models trained on our virtual dataset can be applied in real world. And Pano-YOLO is capable of real-time object detection task on high-resolution 360° panoramic images and videos.