Deep asymmetric extraction and aggregation for infrared small target detection

Recently, many excellent algorithms have made great progress in object detection, but there are also problems in these algorithms’ performance on targets of different sizes, and in particular in small object detection. Aiming at the problem of insufficient feature representation by the feature extractor, in this paper we propose a lightweight algorithm to improve feature extraction. The algorithm includes three modules. First, considering that the shallow features in feature extraction contain much background noise, in this paper we design a multi-level feedback propagation model based on a Gaussian high-pass filter. The shallow layers are enhanced using the filter and then back-propagated to add the upper shallow layer features and obtain new shallow layer features. This process is performed on the newly generated shallow layer for n iterations, which is beneficial for enhancing targets in the foreground area and suppressing background noise. Second, we form a stacked dilated convolution module with different dilation rates to cover the entire deep feature layer densely, which enlarges the receptive field and enriches the contextual information. Finally, we build a multi-scale fusion module to fuse the above-mentioned enhanced shallow and deep features to obtain output features with powerful representational ability for detection tasks. In addition, the model is easily embedded into existing approaches to enhance their performance. We build the model on the VGG-16 and ResNet-50 backbones and successfully applied it on Darknet-19 and Darknet-53 to verify its effectiveness and stability. The experiments on the COCO dataset prove that the proposed algorithm outperforms the state-of-art methods, with a mean average precision improvement reaching 2% on average. The effect is remarkable on small targets and complex backgrounds. Furthermore, it does not affect the detection speed significantly, so real time detection requirements can still be met.

In the field of bearing defect detection, Aiming at the problem of low efficiency in manual inspection and prone to missed detections in scenarios with small target defects, and overlapping targets, an improved YOLOv5-based object detection method is proposed. Firstly, in terms of feature extraction, the C3 modules in the original backbone of YOLOv5 are replaced with the finer-grained Res2Block modules to improve the model’s feature extraction ability. Secondly, in terms of feature fusion, a Bidirectional Feature Pyramid Network (BiFPN) is added to the original neck of YOLOv5 to enhance the fusion ability of shallow graphic features and deep semantic features. Finally, the performance of the improved YOLOv5 algorithm is validated through ablation experiments and comparative experiments with other defect detection algorithms, including the Small_obj algorithm the existing method of adding a small target detection head for identifying small target defects. The experimental results demonstrate that the improved YOLOv5 algorithm exhibits high mAP and accuracy in bearing defect detection, enabling more precise identification the types of small target defects on bearings in complex scenarios with multiple coexisting defects and overlapping detection targets, thereby providing valuable reference for practical bearing defect detection.

The detection of small infrared targets lacking texture and shape information in the presence of complex background clutter is a challenge that has attracted considerable research attention in recent years. Typical deep learning-based target detection methods are designed with deeper network structures, which may lose targets in the deeper layers and cannot directly be used for small infrared target detection. Therefore, we designed the attention fusion feature pyramid network (AFFPN) specifically for small infrared target detection. Specifically, it consists of feature extraction and feature fusion modules. In the feature extraction stage, the global contextual prior information of small targets is first considered in the deep layer of the network using the atrous spatial pyramid pooling module. Subsequently, the spatial location and semantic information features of small infrared targets in the shallow and deep layers are adaptively enhanced by the designed attention fusion module to improve the feature representation capability of the network for targets. Finally, high-performance detection is achieved through the multilayer feature fusion mechanism. Moreover, we performed a comprehensive ablation study to evaluate the effectiveness of each component. The results demonstrate that the proposed method performs better than state-of-the-art methods on a publicly available dataset. Furthermore, AFFPN was deployed on an NVIDIA Jetson AGX Xavier development board and achieved real-time target detection, further advancing practical research and applications in the field of unmanned aerial vehicle infrared search and tracking.

To address the challenges of multi-scale objects, dense distributions, occlusions, and numerous small targets in UAV image detection, we present CMS-YOLOv7, a real-time target detection method based on an enhanced YOLOv7 model. Firstly, the detection layer P2 for small targets was added to YOLOv7 to enhance the detection ability of small and medium-sized targets, and the deep detection head P5 was taken out to mitigate the influence of excessive downsampling on small target images. The anchor frame was calculated by the K-means++ method. Using the concept of Inner-IoU, the Inner-MPDIoU loss function was constructed to control the range of the auxiliary border and improve detection performance. Furthermore, the CARAFE module was introduced to replace traditional upsampling methods, offering improved integration of semantic information during the image upsampling process and enhancing feature mapping accuracy. Simultaneously, during the feature extraction stage, a non-strided convolutional SPD-Conv module was constructed using space-to-depth techniques. This module replaced certain convolutional operations to minimize the loss of fine-grained information and improve the model’s ability to extract features from small targets. Experiments on the UAV aerial photo dataset VisDrone2019 demonstrated that compared with the baseline YOLOv7 object detection algorithm, CMS-YOLOv7 achieved an improvement of 3.5% mAP@0.5, 3.0% mAP@0.5:0.95, and the number of parameters decreased by 18.54 M. The ability of small target detection was significantly enhanced.

Small and dim target detection is a critical challenge in hyperspectral remote sensing, particularly in complex, large-scale scenes where spectral variability across diverse land cover types complicates the detection process. In this paper, we propose a novel target reasoning algorithm named Adaptive Global Dense Nested Reasoning Network (AGDNR). This algorithm integrates spatial, spectral, and domain knowledge to enhance the detection accuracy of small and dim targets in large-scale environments and simultaneously enables reasoning about target categories. The proposed method involves three key innovations. Firstly, we develop a high-dimensional, multi-layer nested U-Net that facilitates cross-layer feature transfer, preserving high-level features of small and dim targets throughout the network. Secondly, we present a novel approach for computing physicochemical parameters, which enhances the spectral characteristics of targets while minimizing environmental interference. Thirdly, we construct a geographic knowledge graph that incorporates both target and environmental information, enabling global target reasoning and more effective detection of small targets across large-scale scenes. Experimental results on three challenging datasets show that our method outperforms state-of-the-art approaches in detection accuracy and achieves successful classification of different small targets. Consequently, the proposed method offers a robust solution for the precise detection of hyperspectral small targets in large-scale scenarios.

Image content variation is a typical and challenging problem in no-reference image-quality assessment (NR-IQA). This work pays special attention to the impact of image content variation on NR-IQA methods. To better analyze this impact, we focus on blur-dominated distortions to exclude the impacts of distortion-type variations. We empirically show that current NR-IQA methods are inconsistent with human visual perception when predicting the relative quality of image pairs with different image contents. In view of deep semantic features of pretrained image classification neural networks always containing discriminative image content information, we put forward a new NR-IQA method based on semantic feature aggregation (SFA) to alleviate the impact of image content variation. Specifically, instead of resizing the image, we first crop multiple overlapping patches over the entire distorted image to avoid introducing geometric deformations. Then, according to an adaptive layer selection procedure, we extract deep semantic features by leveraging the power of a pretrained image classification model for its inherent content-aware property. After that, the local patch features are aggregated using several statistical structures. Finally, a linear regression model is trained for mapping the aggregated global features to image-quality scores. The proposed method, SFA, is compared with nine representative blur-specific NR-IQA methods, two general-purpose NR-IQA methods, and two extra full-reference IQA methods on Gaussian blur images (with and without Gaussian noise/JPEG compression) and realistic blur images from multiple databases, including LIVE, TID2008, TID2013, MLIVE1, MLIVE2, BID, and CLIVE. Experimental results show that SFA is superior to the state-of-the-art NR methods on all seven databases. It is also verified that deep semantic features play a crucial role in addressing image content variation, and this provides a new perspective for NR-IQA.

Dim and small infrared target fast detection guided by visual saliency

Research on Infrared Dim and Small Target Detection Algorithm Based on Low-Rank Tensor Recovery

To detect small infrared targets under the condition of dense clutters, we propose a single-frame target detection algorithm based on a small bounding-box filter, which is characterized by good adaptability to the position and size of a small target. During the small target detection process, the proposed algorithm first searches for the local maximum gray pixel and then, a set of concentric bounding boxes whose center is the pixel found in the first step is constructed, and the detection thresholds of a neighboring region of this pixel are calculated based on the bounding boxes. Finally, the minimum threshold is used to detect small target pixels in the neighboring region. A fast version of the proposed algorithm is a minimum bounding-box filter, which can be implemented by dividing an image into blocks and using the mid-range and range to assess the concentration trend and dispersion of the background. Simulation and analysis results show that the proposed algorithm can achieve high detection probability and low false alarm rate when detecting small targets in the complex background; while its fast version has high computational efficiency. The proposed algorithm can be used in infrared searching and tracking systems.

Most existing target detection models are designed for image targets in natural scenarios and the detection results of small targets in aerial images are less than satisfactory. This paper tries to solve the locating and classification problem of small targets in aerial images from the perspectives of fusion between shallow features and deep features, receptive field enhancement, and a lightweight feature extraction network. To realize accurate detection of small targets in aerial images, on the basis of SSD (Single Shot MultiBox Detector), an aerial image target detection model MF-SSD based on multi-scale feature fusion and receptive field enhancement is proposed. In this model, a feature fusion sub-network is established, in which, the two-level fusion of shallow features and skip-connection fusion of intermediate features are provided. In addition, a multi-branch Inception structure D-Inception based on dilated convolution is utilized to achieve receptive field enhancement, which enables the system to better extract target features and adapt to the scale change of target. To address the increased computational complexity of the model after feature fusion and receptive field enhancement processing, a lightweight convolutional neural network based on Mobilenet is used to replace the basic network, which can not only reduce the complexity of the network model but also maintain high detection precision. The experimental results show that the target detection model MF-SSD can be used to effectively detect small targets in aerial images.KeywordsSmall target detectionFeature fusionReceptive field enhancementLightweight convolutional neural network (CNN)

Various forms of social interactions are often char-acterized by toxic or offensive words that can be collectively referred to as offensive languages, which has become a unique linguistic phenomenon in social media platforms. How to detect and identify these offensive languages in social media platforms has become one of the important research in the field of natural language processing. Existing methods utilize machine learning algorithms or text representation models based on deep learning to learn the features of offensive languages and identify them, which have achieved good performances. However, traditional machine learning-based methods mainly rely on keyword identi-fication and blocking, deep learning-based methods do not ade-quately explore the fused deep semantic features of the content by combining word-level embeddings and sentence-level deep semantic feature representations of sentences, which cannot ef-fectively identify offensive languages that do not contain common offensive words but indicate offensive meanings. In this research, we propose a novel offensive language identification model based on deep semantic feature fusion, which uses the pre-trained model Bert to obtain word-level embedding representations of offensive languages, and then integrates the RCNN that combines with the attention mechanism to extract the fused deep semantic feature representations of offensive languages, and label encoder and offensive predictor to improve the identification accuracy and generalization ability of the model so that the performances of the model do not rely on the offensive language lexicon entirely and can identify offensive languages that do not contain common offensive words but indicate offensive meanings. Experimental results on Wikipedia and Twitter comment datasets show that our proposed model can better understand the context and discover potential offensive meanings, and outperforms existing methods.

Accurate detection of fresh jujube fruits plays a vital role in precision agriculture, enabling reliable yield estimation and supporting automation tasks such as robotic harvesting. To address the challenges of detecting such small targets (≤32 × 32 pixels) in complex orchard environments, this study proposes JFST-DETR, an efficient and robust detection model based on the Real-Time DEtection TRansformer (RT-DETR). First, to address the insufficient feature representation for small jujube fruit targets, a novel module called the Global Awareness Adaptive Module (GAAM) is designed. Building on GAAM and the innovative Spatial Coding Module (SCM), a new Spatial Enhancement Pyramid Network (SEPN) is proposed. Through the spatial-depth transformation domain and global awareness adaptive processing units, SEPN captures fine-grained features of small targets, enhancing the detection accuracy for small objects. Second, a Dynamic Sampling (DySample) operator is adopted, which optimizes feature space details via dynamic offset calculation and lightweight design, improving detection accuracy while reducing computational costs. Finally, to solve the problem of complex background interference caused by foliage occlusion and illumination variations, Pinwheel-Shaped Convolution (PSConv) is introduced. By using asymmetric padding and multi-directional convolution, PSConv enhances the robustness of feature extraction, ensuring reliable recognition in complex agricultural environments. Experimental results show that JFST-DETR achieves precision, recall, F1, mAP@50, and mAP@50:95 of 93%, 86.8%, 89.8%, 94.3%, and 75.2%. Compared to the baseline model, these metrics improve by 0.8%, 3.7%, 2.4%, 2.6%, and 3.1%, respectively. Cross-dataset evaluations further confirm its strong generalizability, demonstrating potential as a practical solution for small-target detection in intelligent horticulture.

Most existing small object detection methods rely on residual blocks to process deep feature maps. However, these residual blocks, composed of multiple large-kernel convolution layers, incur high computational costs and contain redundant information, which makes it difficult to improve detection performance for small objects. To address this, we designed an improved feature pyramid network called L Feature Pyramid Network (L-FPN), which optimizes the allocation of computational resources for small object detection by reconstructing the original FPN structure. Based on L-FPN, we further proposed a small object detector named BPD-YOLO. We introduce a Dual-phase Asymptotic Feature Fusion mechanism (DAFF), where the shallow and deep semantic features extracted from the backbone network are initially fused in parallel to mitigate the semantic gap. Subsequently, the intermediate semantic layers are progressively integrated, enabling effective fusion of both shallow and deep feature representations. Additionally, we designed the Deep Spatial Pyramid Fusion module (DSPF), which generates multi-scale feature representations as an alternative to conventional residual block stacking, thereby reducing computational overhead. In the shallow feature extraction stage, DSPF focuses on semantic integration and enhances the extraction of small object features. This strategy, which adaptively selects different modules based on the resolution of the feature maps, is referred to as the Decoupled feature Extraction-semantic Integration mechanism (DEI). Finally, we conducted extensive experiments and thorough evaluations on both the VisDrone and TinyPerson datasets. The results demonstrate that, on the VisDrone dataset, compared to the baseline model YOLOv8n + p2, our BPD-YOLO model with L-FPN achieves a 2.8% improvement in mAP50 and a 1.4% increase in mAP50-95. On the TinyPerson dataset, BPD-YOLO further demonstrates its superiority in high-resolution feature extraction, effectively enhancing detection accuracy while significantly reducing computational costs.

The infrared image in the sea background usually contains random noise, sea waves, fish scale light and other clutter. Sea waves and fish scale light interference are similar to small infrared targets, so it is difficult to eliminate such interference and detect small infrared targets quickly and accurately. In this paper, a small target detection method based on non-local similarity is proposed. The interference is eliminated by using the class feature similarity of interference and the overall feature uniqueness of the small target. Firstly, this paper performs filtering to remove random noise, and the suspected small targets can be extracted by executing Top Hat filtering. Secondly, a multi-feature vector is constructed for all suspected targets. Finally, this paper analyzes the characteristics of different interference and executes clustering step by step. The wave edge interference and fish scale light interference can be eliminated in turn. Feature threshold is set based on the interference that has been identified to obtain the true small infrared target according to the uniqueness of the characteristics of the true target. The experimental results show that compared with other small target detection algorithms, the proposed algorithm can detect small infrared targets in complex sea background effectively.

Asymmetric double-winged multi-view clustering network for exploring diverse and consistent information