Search

The key to accurate perception and efficient decision making of autonomous driving is the long-range detection of traffic signs. Long-range detection of traffic signs has the problems of small traffic sign size and complex background. In order to solve these problems, this paper proposes a lightweight model for traffic sign detection based on real-time detection transformer (TSD-DETR). Firstly, the feature extraction module is constructed using multiple types of convolutional modules. The model extracts multi-scale features of different levels to enhance feature extraction ability. Then, small object detection module and detection head are designed to extract and detect shallow features. It can improve the detection of small traffic signs. Finally, Efficient Multi-Scale Attention is introduced to adjust the channel weights. It aggregates the output features of three parallel branches interactively. TSD-DETR achieves a mean average precision (mAp) of 96.8% on Tsinghua-Tencent 100K dataset. It is improved by 2.5% compared with real-time detection transformer. In small object detection, mAp improved by 9%. TSD-DETR achieves 99.4% mAp on the Changsha University of Science and Technology Chinese Traffic Sign Detection Benchmark dataset, with an improvement of 0.6%. The experimental results show that TSD-DETR reduces the number of parameters by 9.06M by optimizing the model structure. On the premise of ensuring the real-time performance of the model, the detection accuracy of the model is improved greatly. The results of ablation experiments show that the feature extraction module and small object detection module proposed in this paper are conducive to improving the detection accuracy. [ABSTRACT FROM AUTHOR]

With the advance of automation in constructing simply supported box girders for high-speed railways, traditional manual binding of steel mesh has been transmitted into automated welding. Nevertheless, dimensional quality inspection of steel mesh is still stagnant in manual operation, which is highly cost-ineffective and time-consuming. Current research lacks real-time and accurate detection methods for super-large steel mesh up to 32 m × 10 m. This paper proposes an efficient and precise dimensional quality inspection method for super-large steel mesh. Firstly, an automated image acquisition framework based on customized devices and method is proposed to synchronize equipment processing and inspection. The geometric features of the steel bar spacing are then explored, leading to the design of a fixed-spacing deformable attention module for detecting rebar images. Additionally, a "tic-tac-toe" template, replacing the traditional "discrete circle" template, is designed to enhance the accuracy of oriented fast and rotated brief. Experimental validation indicates a 0.156 improvement in the average precision of the fixed-spacing deformable attention module and a 39.6% increase in the accuracy of oriented fast and rotated brief. Field testing in a high-speed railway girder yard in China demonstrates second-level efficiency and millimeter-level accuracy of the proposed method. This study improves the deformable detection transformer, oriented fast, rotated brief and introduces a dimensional quality inspection method suitable for super-large steel mesh. It contributes significantly to artificial intelligence via replacing manual labor, promoting automation in high-speed railway construction, and thus benefiting practical engineering. • Develop an automated dimensional quality inspection method for steel mesh with size up to 32 m × 10 m. • Realize concurrent equipment processing and inspection by customized image collection device and method. • Propose fixed-spacing detection transformer with 0.156 increase in average precision. • Improve oriented fast and rotated brief by "tic-tac-toe" template for 39.6% improvement in accuracy. • Conduct field testing on a high-speed railway girder yard in Guangdong, Ghina. [ABSTRACT FROM AUTHOR]

Fire prevention in industrial settings are paramount for ensuring human safety and economic stability. However, current mainstream DETR detectors face significant challenges in applications due to the necessity for extensive memory accesses and inference delays. To tackle these complexities, we propose an Industrial Fire Smoke Detector based on an end-to-end structured framework. In a series of innovative optimizations, we firstly adopt a more lightweight backbone network, LeanNet, for feature extraction. Combined with the optimized Transformer architecture, this approach enhances the model's detection speed to address real-time challenges effectively. Secondly, we introduce a Feature Fusion Network based on alignment mechanisms to enhance the DETR model's multi-scale object representation capabilities without significantly increasing latency. Subsequently, to facilitate easier training and optimization of IFS-DETR, we introduce IoU-aware query selection and an aspect ratio-based denoising training strategy, and enhance the localization loss function using Inner-SIoU. Finally, we deploy IFS-DETR on NVIDIA Jetson Orin Nano. The dataset is available at https://github.com/Sonnenb1ume/IFS-DETR. • LeanNet, an efficient backbone network based on depth-separable convolution. • AFFNet, a feature fusion network based on alignment mechanism and EFCA attention. • Improve training efficiency with more effective training strategies and box losses. • Deployed IFS-DETR in edge devices for real-time applications. [ABSTRACT FROM AUTHOR]

Recently DEtection TRansformer(DETR)-based models obtain remarkable performance in object detection and various foundational vision tasks. However, its performance is impeded by high computational demands since it exhibits quadratic scaling with the number of feature tokens. To mitigate redundant computations in some areas like the background, existing works propose static token selection methods, which choose a predefined portion of tokens to forward. However, it is intuitive that the complexity of inference for detection tasks varies depending on the input images. Static token selection methods rely on a fixed keeping ratio, causing performance degradation in complex scenes and inefficiency in simple scenes. To address this issue, we propose an A daptive T oken S election method for DETR (ATS-DETR) that dynamically chooses the token keeping ratio based on the complexity of the input to retain the most salient tokens. To explicitly control the sparsity and improve the performance of ATS-DETR, we put forward a novel approach called Dual Teacher Supervision to train the ATS-DETR. Specifically, we utilize a weak teacher to assist the model in distinguishing input complexity and a strong teacher for enhancing overall model performance through feature distillation. We further introduce the Global Distillation to minish the disparities of the feature patterns extracted from ATS-DETR and the strong teacher model. Extensive experiments demonstrate that ATS-DETR attains better performance compared to Deformable DETR while achieving an 83% reduction of GFLOPs in the encoder, and outperforms all the static token selection methods. • Methodology for adaptive token selection in DETR based on input difficulty. • An input-aware adaptive training strategy with dual teacher supervision. • A distillation approach to minish the disparities in the feature patterns extracted from adaptive and static model. • Adaptive token selection demonstrates superiority over static token selection methods. [ABSTRACT FROM AUTHOR]

When a ship navigates in ice-covered areas, its hull experiences motion and vibration responses due to ship-ice contact. The magnitude of such responses can be represented well by acceleration, which is easy to measure onboard ice-going ships. The level of acceleration is determined by the environmental and operational conditions through excitation via the ship-ice contact. Conversely, it is possible to infer excitation and environmental conditions using acceleration measurements via appropriate methods. Accelerations associated with such responses can also affect the functionality of the onboard equipment and therefore, should be quantified. Nevertheless, studies on shipborne acceleration measurements in ice fields and their links to prevailing excitations and environmental conditions are rare. This study explores the feasibility of using shipborne acceleration measurements to infer the corresponding excitations (i.e., ship-ice contacts) as a first step towards the inference of ice conditions via acceleration measurements. An automatic identification method for ship-ice contact is proposed that can effectively separate ship-ice contact events from acceleration measurement signals. The acceleration signals are decomposed in the time-frequency map obtained through a short-time Fourier transformation (STFT). An improved detection transformer (DETR) model was used to automatically identify ship-ice contact events and achieved good accuracy in the validation set. This model was then applied to automatically identify ship-ice contacts using acceleration signals under various environmental and operational conditions. The frequency of the identified ship-ice contact reasonably reflects its dependence on prevailing environmental and operational conditions, which confirms the validity of the proposed methodology. • A novel concept is proposed to identify ship-ice contacts from onboard acceleration measurements. • An improved DETR model is applied to the time-frequency map obtained by STFT. • The DETR model is applied to various ice conditions and the ship-ice contact frequencies is analyzed. • Study quantifies ship speed and ice concentration impact on ship-ice contact frequency. [ABSTRACT FROM AUTHOR]

Transformer-based detectors have recently achieved remarkable success in object detection, revolutionizing the field with their efficiency and accuracy. However, applying these models to high-resolution images presents significant challenges due to the increased computational demands and complexity of processing dense, high-resolution data. In this paper, we introduce a novel model specifically designed for object detection from high-resolution imagery. This model incorporates a multi-layer object-focus network along with a transformer encoder-decoder structure. Specifically, the model employs a dual-head strategy in the object-focus network to balance the detailed analysis of small objects with computational efficiency. This is achieved by leveraging data sparsity to reduce unnecessary computations in massive background regions. Additionally, to improve detection performance for small objects, we propose a method to effectively apply the transformer encoder-decoder structure combined with the object-focus network on the multi-layer feature maps of the feature pyramid. Our extensive evaluations of the VisDrone, MS-COCO, and UAVid datasets demonstrate that our model outperforms other DETR-based detectors in both detection accuracy and computational speed, highlighting its superior performance. These results indicate a significant advancement in the field of high-resolution object detection utilizing transformer architecture. [ABSTRACT FROM AUTHOR]

• Our research mainly focuses on the detection of multiple different categories of targets and defects in transmission lines, and considers small targets and occluded targets. • Different from the traditional tandem object detection paradigm, a new cascade object detection model structure is designed by us: cascade distillation structure. • The components inside the cascade modules are specifically designed to obtain efficient, high-quality information. • Extract features through multiple spatial display interactions and control information redundancy through a lightweight gated channel attention mechanism. • The detection box accuracy is optimized by aggregating contextual information. The cascade distillation detection model is proposed to address the problems of low accuracy of multiple small target defect detection and lack of robustness under complex environments in transmission line UAV intelligent inspection. Firstly, the cascading modules of the shared backbone network are designed: the dynamic anchor box generation module and the bounding box refinement module. The former collects and processes areas where there may be targets, outputs high-quality location information, and distills it to the latter, and the latter is responsible for refining and filtering this part of the information, which solves the problems of missed detection of model targets and lack of a priori information on target locations. Secondly, a high-order aggregated feature extraction backbone network is proposed to enhance the feature extraction of small and medium-sized targets and obscured targets on transmission lines. Then, a spatially deformable attention mechanism is designed to optimize the selection of sampling points for transmission line targets and defects and the calculation of their attention weights by aggregating contextual information in the vicinity of the targets, enhancing the module's ability to refine and adjust the bounding boxes. Finally, a large dataset of transmission line defect targets is constructed and generated, and the validity of the model is verified by experiments. [ABSTRACT FROM AUTHOR]