Small Target Detection Algorithm Research Articles

An Improved Lightweight Small Target Detection Algorithm for Air-to-Ground Image

An Improved Lightweight Small Target Detection Algorithm for Air-to-Ground Image

Infrared small target detection algorithm based on two-dimensional modal decomposition combined with patch contrast

Infrared small target detection algorithm based on two-dimensional modal decomposition combined with patch contrast

Lightweight Infrared Small-Target Detection Algorithm

Lightweight Infrared Small-Target Detection Algorithm

Infrared small target detection algorithm based on spatial dissimilarity weighted local contrast

To solve the problem that the local contrast algorithm is easily influenced by the heavy clutter background or strong noise and does not fully consider the spatial neighbourhood characteristics of the small target, a spatial dissimilarity weighted local contrast-based method (SDWLCM) for infrared small target detection is proposed in this paper. Firstly, the two-dimensional difference of the Gaussian filter with central excitation and lateral suppression characteristics is chosen to preprocess the original infrared image for removing the flat background and improving the signal-to-noise ratio (SNR) of the small target. Secondly, the spatial dissimilarity between the target and its surrounding backgrounds is designed for local contrast weighting to generate the contrast saliency map so that the heavy clutter background is greatly suppressed and the small target is further highlighted. Thirdly, the saliency map is segmented by the adaptive threshold to get the real targets. Experimental results show that, compared with other methods, the SDWLCM, which owns not only a higher SNR gain and a larger background suppression factor but also a higher detection rate and a lower false alarm rate, is confirmed to be an effective method to detect the small targets.

Unmanned Aerial Vehicle Recognition of Maritime Small-Target Based on Biological Eagle-Eye Vision Adaptation Mechanism

Inspired by the background adaptive mechanism of eagle vision in different hunting environment, a biological eagle-eye vision adaptation mechanism algorithm is proposed for unmanned aerial vehicle (UAV) to detect the long-distance maritime small target in complex and changeable sea environment. First, the various environment adjustment abilities are summarized according to the physiological structures and characteristics of eagle vision. Second, the mathematical models of glaring adaptation, dim adaptation, color adaptation, and the background adaptation are established based on the background adaptation mechanisms of eagle vision. Last but not least, our proposed biological eagle-eye vision adaptation method and other five comparative experiments are implemented in different scenes, such as dazzling, cloudy, dusk, etc. The results of various evaluation indices show that the information of maritime small target is retained satisfactorily and the background of sea or sky is restrained effectively by the proposed algorithm. The maritime small target detection algorithm can be used for the vision system of UAV operating on the sea. It provides a feasible solution for UAV's remote vision autonomous navigation in changeable sea environment.

Research on Traffic Sign Detection Algorithm Based on Improved YOLOv3

As the traffic sign target detection has the defects of small size, low resolution and unobvious features, an improved YOLOv3 network model is proposed. Firstly, using a K-means algorithm TT100K traffic sign data set clustering analysis and redefine the initial candidate frame of the network. Secondly, this paper improves the FPN structure in the original network, retain the large-scale prediction of 52×52 in the original network, and then use the feature map output from the second down-sampling in the YOLOv3 network to establish a larger-scale prediction of 108×108. For the purpose of solving the size of image and distortion problems, the pyramid pooling operation with fixed block sizes of 5, 9, 13 is used before the detection layer, and then the output features are merged with the original feature map, so as to achieve the same size for inputs of different sizes output. We use the improved YOLOv3 network model with its original model and other small target detection algorithms to conduct comparative experiments on the TT100K data set. The results show that the improved YOLOv3 network model can detect traffic signs more effectively, with better detection accuracy and real-time performance.

Infrared small target detection method based on multi-scale feature fusion

Infrared small target detection is widely used in aerial target detection and tracking system because of its long detection distance and strong anti-jamming ability. Aiming at the shortcomings of the current infrared small target detection algorithm, such as low precision rate and high false alarm rate when dealing with complex background. In this paper, we propose an end-to-end infrared small target detection model (called MFSSD) based on multi-scale feature fusion. Considering the trait of the targets, we propose a feature fusion module by using a refinement and fusion feature map method, and improve the correlation of different channels through the SP Module. The experimental results of three different sequences of infrared image detection show that the average detection accuracy of the MFSSD algorithm in infrared small target detection is as high as 87.8%. Compared with the traditional multi-scale feature fusion detection algorithm, both the precision rate and the recall rate have been significantly improved.

Infrared small target detection based on the dynamic particle swarm optimization

In this paper, a new method is introduced for detecting small targets in infrared images. The proposed method is based on the particle swarm optimization algorithm (PSO). Considering the nature of the small target detection problem, a dynamic optimization algorithm is developed to detect targets. The proposed algorithm is called the dynamic particle swarm detector (DPSD). Unlike common small target detection algorithms, the computationalcomplexityoftheproposedalgorithmis not linear to number of pixels in the input image. Therefore, it is capable of operating on a large-scale image in a relatively small timeslot. Many experiments are carried out to evaluate the effectiveness of the proposed method, where the DPSD is compared to seven well-known methods in terms of quantitative detection metrics. The results show that the proposed detector outperforms the baseline algorithms.

A small target detection algorithm in infrared image by combining multi-response fusion and local contrast enhancement

The detection of small target in infrared (IR) images is the key technology for IR search, tracking and early warning. When the target locates at distance, the target occupies tiny area, shows low signal-to-noise ratio, which is hard to be detected. An infrared small target detection algorithm by combining multi-response fusion and local contrast enhancement (MRF-LCE) is introduced in this paper. First, Laplacian of Gaussian (LoG) filter and negative LoG filter are utilized to obtain the response map of bright and dark targets, respectively. Second, a multi-response fusion method is designed to respond to the bright and dark targets, simultaneously. Finally, a local contrast method is implemented to further enhance the targets and suppress the background, and these small targets are segmented by an adaptive threshold method. Experiments on three different real sequences show that the proposed algorithm can effectively detect targets in different scenarios.

Infrared maritime dim small target detection based on spatiotemporal cues and directional morphological filtering

The existing small target detection algorithms will suffer severe performance deterioration when the weak target is submerged in heavy cluttered infrared (IR) maritime image. To settle this problem, a small target detection scheme is developed based on spatiotemporal cues and multidirectional morphological filtering. Initially, considering the small target will be isotropic because of the point spread function of the long-distance thermal imaging, while background edge clutters are generally local directional, the multidirectional structuring elements (MSEs) are constructed. By incorporating constructed MSEs and morphological operations, the directional morphological filtering (DMF) is established to measure the multidirectional differences between target region and local surroundings. Then, according to the brightness and isotropic Gaussian-like shape of small target in local spatial domain, the local intensity difference measure (LIDM) and the local Gaussian-like shape measure (LGSM) are presented to differentiate the small target from background clutters. After that, in order to integrate the two spatial properties, the multidirectional improved top-hat filter (MITHF) is proposed by fusing LIDM and LGSM to highlight potential small targets and suppress strong structural edge clutters. Finally, considering the consistency of targets and the fluctuation of sea clutters in temporal domain, a new appearance descriptor namely histogram of oriented morphological filtering (HOMF) is designed to characterize the small target by fully exploiting the advantages of DMF, and the HOMF-based multi-frame confirmation strategy (HMCS) is developed to further identify real small targets and eliminate false alarms. Extensive experiments prove that proposed algorithm is superior to the compared methods, especially for dim small target buried in intricate maritime background clutter.

Three-dimensional pipeline Hough transform for small target detection

We propose a 3D Hough transform (3D-HT) algorithm that can overcome the disadvantages of high complexity and large data storage space of the existing 3D-HT-based small target detection algorithms. The proposed algorithm uses two coordinates at different times and coordinate errors to construct a 3D pipeline. Subsequently, it counts the number of points in the 3D pipeline and confirms the presence of a target trajectory in the pipeline when the number of points exceeds a predefined threshold. Finally, it performs trajectory merging and filtering before outputting the target trajectory coordinates. The proposed algorithm has low complexity because the used trajectory parameters are the coordinates in the data space, and only a linear transform between the coordinates is required. Unlike the existing algorithms that use an accumulator array to represent the Hough space, the proposed algorithm uses only a single position-adaptive cumulative cell in the Hough space. Therefore, there is no limitation on data storage in the Hough space. Simulation and analysis show that the small target detection algorithm based on the proposed transform is robust to noise, requires small data storage space, and has high computational efficiency. The proposed algorithm can be used in infrared and radar small target trajectory detection systems.

RISTDnet: Robust Infrared Small Target Detection Network

The infrared (IR) small target detection algorithm with a high detection rate, low false alarm rate, and high real-time performance has significant application value in the field of IR remote sensing. IR small targets in complex backgrounds have low contrast and low signal-to-noise ratio (SNR). Therefore, small target detection is more difficult. Traditional IR small target detection is generally implemented by local contrast methods (LCM), nonlocal autocorrelation methods (NAM), and adaptive segmentation. In this letter, a robust infrared small target detection network (RISTDnet) is proposed based on deep learning. In RISTDnet, a feature extraction framework combining handcrafted feature methods and convolutional neural networks is constructed, a mapping network between feature maps and the likelihood of small targets in the image is established, and a threshold is applied on the likelihood map to segment real targets. Experimental results show that the RISTDnet can detect small targets with different sizes and low SNRs in complex backgrounds and have better effectiveness and robustness against existing algorithms.

A Traffic-Sign Detection Algorithm Based on Improved Sparse R-cnn

Automatic traffic-sign detection is a hot topic in computer vision and one of the critical technologies of intelligent transportation. The Transformer structure has recently become a research hotspot due to its excellent performance. We hope to apply this structure to the design of traffic sign detection algorithms. Therefore, we make some improvements to Sparse R-cnn, a neural network model inspired by Transformer. Sparse R-cnn is a novel model, and its core idea is to replace hundreds of thousands of candidate anchors in the RPN network with a small set of proposal boxes. The experiments in our paper proved that the performance of the Sparse R-cnn model is better than other existing general object detection models. Based on the original Sparse R-cnn inspiration, an improved Sparse R-cnn model is proposed. First, a novel backbone for the task of traffic-sign detection is proposed. Multi-scale fusion structure is the essential method of improving the algorithm for small target detection, so improving the multi-scale capability of the backbone is a required method for designing traffic sign detection. So, we made further improvements to the existing backbone ResNest. We enhanced the multi-scale representation ability of the backbone by constructing hierarchical residual-like connections within each single radix block in the original ResNest. We call the improved backbone Res2Nest. The novel backbone proposed by us shows better performance without introducing excessive computational costs to the model. In addition, the attention mechanism is also an effective method to improve the detection of traffic signs, so we set up a branch network for recalibrating the channel feature response adaptively through the Global Average Pooling (GAP) operation and a fully connected layer. It can also be seen as the implementation of the cross-channel self-attention mechanism. After experiments by TT100K dataset, our method would attain a better accuracy and robustness.

Fast and Robust Infrared Image Small Target Detection Based on the Convolution of Layered Gradient Kernel

Infrared (IR) small target detection is challenging because the IR imaging lacks detailed features, weak shape features, and a low signal-to-noise ratio (SNR). The existing small IR target detection methods usually focus on improving their high detective performance without considering the execution time. However, high-speed detection is vital for various applications, such as early warning systems, military surveillance, infrared search and track (IRST), etc. This paper proposes a fast and robust single-frame IR small target detection algorithm with a low computational cost while maintaining excellent detection performance. We propose a layered gradient kernel (LGK) based on the contrast properties of the human visual system (HVS) and model it through a three-layer patch image model. The layered gradient kernel is used to convolute with the input IR frame to obtain its gradient map. The target detection is further performed on the acquired gradient map with an adaptive threshold method. This method is compared with eight representative small target detection algorithms to evaluate the performance. Experimental results demonstrate that the algorithm is fast and suitable for real-time applications, and it is very effective even when the small target size is as small as 2×2.

Retracted] Small Target Detection Algorithm Based on Transfer Learning and Deep Separable Network

With the development of deep learning, target detection from vision sensor has achieved high accuracy and efficiency. However, small target detection remains a challenge due to inadequate use of semantic information and detailed texture information of underlying features. To solve the above problems, this paper proposes a small target detection algorithm based on Mask R‐CNN model which integrates transfer learning and deep separable network. Firstly, the feature pyramid fusion structure is introduced to enhance the learning effect of low‐level and high‐level features, especially to strengthen the information channel of low‐level feature and meanwhile optimize the feature information of small target. Secondly, the ELU function is used as the activation function to solve the problem that the original activation function disappears in the negative half axis gradient. Finally, a new loss function F‐Softmax combined with Focal Loss was adopted to solve the imbalance of positive and negative sample proportions. In this paper, self‐made data set is used to carry out experiments, and the experimental results show that the proposed algorithm makes the detection accuracy of small targets reach 66.5%.

Investigation of infrared dim and small target detection algorithm based on the visual saliency feature

Infrared dim and small target detection has an important role in the infrared thermal imaging seeker, infrared search and tracking system, space-based infrared system and other applications. Inspired by human visual system (HVS), based on the fusion of significant features of targets, the present study proposes an infrared dim and small target detection algorithm for complex backgrounds. Firstly, in order to calculate the target saliency map, the proposed algorithm initially uses the difference of Gaussian (DoG) and the contourlet filters for the preprocessing and fusion, respectively. Then the multi-scale improved local contrast measure (ILCM) method is applied to obtain the interested target area, effectively suppress the background clutter and improve the target signal-to-clutter ratio. Secondly, the optical flow method is used to estimate motion regions in the saliency map, which matches with the interested target region to achieve the initial target screening. Finally, in order to reduce the false alarm rate, forward pipeline filtering and optical flow estimation information are applied to achieve the multi-frame target recognition and achieve continuous detection of dim and small targets in image sequences. Experimental results show that compared with the conventional Tophat (TOP-HAT) and ILCM algorithms, the proposed algorithm can achieve stable, continuous and adaptive target detection for multiple backgrounds. The area under curve (AUC) and the harmonic average measure F1 are used to measure the overall performance and optimal performance of the target detection effect. For sky, sea and ground backgrounds, the test results of proposed algorithm for most sequences are 1. It is concluded that the proposed algorithm significantly improves the detection effect.

CNN‐based infrared dim small target detection algorithm using target‐oriented shallow‐deep features and effective small anchor

CNN‐based infrared dim small target detection algorithm using target‐oriented shallow‐deep features and effective small anchor

Dim and Small Targets Detection in Sequence Images Based on Spatiotemporal Motion Characteristics

In order to effectively enhance the low detection rates of dim and small targets caused by dynamic backgrounds, this paper proposes a detection algorithm for dim and small targets in sequence images based on spatiotemporal motion characteristics. With regard to the spatial domain, this paper proposes an improved anisotropic background filtering algorithm that makes full use of the gradient differences between the target and the background pixels in the eight directions of the spatial domain and selects the mean value of the three directions with the lowest diffusion function in the eight directions as the differential filter to obtain a differential image. Then, the paper proposes a directional energy correlation enhancement algorithm in the time domain. Finally, based on the above preprocessing operations, we construct a dim and small targets detection algorithm in sequence images with local motion characteristics, which achieves target detection by determining the number of occurrences of the target, the number of displacements, and the total cumulative area in these sequential images. Experiments show that the proposed detection algorithm in this paper can effectively improve the detection of dim and small targets in dynamic scenes.

Improved RetinaNet model for the application of small target detection in the aerial images

The target detection algorithm for accurately locating the location of objects in the image and their categories is widely used in intelligent traffic, forest fire prevention, electric power patrol and other fields. Existing target detection algorithms detect objects in head-up images, and their accuracy is not high when used in aerial image detection. To solve this problem, a small target detection algorithm based on improved RetinaNet model is proposed. First, replace the ResNet101 deep residual network in the original network with ResNet152, Independent of image features, The model is accelerated by merging ScaleNet network structure; Besides, in order to increase the feeling of the small target detection, to upgrade the existing FPN network output, increase the P2 feature layer, improved the range makes the network overall receptive field and the robustness of small target detection. Through the experimental result shows that compared with the original RetinaNet algorithm, The performance index of the improved algorithm is obviously improved. the average detection accuracy of the improved algorithm for small target detection is quite high 6.21%, it has a good detection effect in many scenarios.

Fast and robust small infrared target detection using absolute directional mean difference algorithm

Infrared small target detection in an infrared search and track (IRST) system is a challenging task. This situation becomes more complicated when high gray-intensity structural backgrounds appear in the field of view (FoV) of the infrared seeker. While the majority of the infrared small target detection algorithms neglect directional information, in this paper, a directional approach is presented to suppress structural backgrounds and develop a more effective detection algorithm. To this end, a similar concept to the average absolute gray difference (AAGD) is utilized to construct a novel directional small target detection algorithm called absolute directional mean difference (ADMD). Also, an efficient implementation procedure is presented for the proposed algorithm. The proposed algorithm effectively enhances the target area and eliminates background clutter. Simulation results on real infrared images prove the significant effectiveness of the proposed algorithm.