Weak light environment poses challenges for target detection. It is difficult for the existing target detection algorithms to track the target in the noise-flooded image. In order to improve the accuracy of the target detection algorithm, it is necessary to process the low-light image first. In this paper, a time-domain filter is used, which combines the dynamic detection of time-domain filter and the denoising ability of spatio-temporal filter to provide help for target detection under low light conditions. This paper focuses on the motion compensation filtering by calculating the associated parameters of the front and back frames through multi-direction edge filtering. The down-sampling differential pyramid is used to avoid the discontinuity of the partition blocks. A convolutional strong edge detection method based on multi-directional template is proposed to improve the edge detection function of traditional spatial filtering template. Finally, kernel correlation filtering (KCF) is applied to the image to realize target detection. By comparing the target detection rate and accuracy before and after filtering, it is proved that the algorithm can effectively improve the detection ability of low-light level image. By comparing with other algorithms, it is proved that the proposed algorithm has more advantages than the uncombined traditional algorithm.
Background estimation and motion target detection under moving platform are key technologies in motion camera system, which has attracted much attention in recent years and is widely used in security monitoring, UAV detection and other optoelectronic detection systems. When the camera is mounted on a moving platform, the movement of the platform causes a corresponding displacement of the image background captured by the camera. Therefore, the stationary background estimation method based on a stationary camera is no longer applicable in this case. In this paper, we propose a method for calculating the image motion information using the dynamic template phase matching method. Homologous signal trajectories are then established based on the image motion information to compensate for the inverse motion of the background image. Finally, a more accurate stationary background image is obtained using the background modeling method.
Moving target detection is always a hot research field, which is widely used in many fields, especially in military field. In this paper, we use the gyroscope data to quickly screen the moving objects under the motion camera. Firstly, the feature points are extracted according to the curvature feature of the point target. Then, the camera is calibrated to obtain the external and internal parameters of the camera, and the gyroscope error is corrected according to the original dense background track recorded by fitting the gyroscope data, and the obtained correction value is compensated back to the original image coordinates of the feature points, to obtain the corresponding world coordinates. Then the point correlation of the feature points in the world coordinate system is carried out to form a stable track. Finally, according to the motion characteristics of the feature points, the moving target trajectory is selected to realize the detection of the moving target under the motion camera. According to the experiments in this paper, the feasibility and high efficiency of the gyroscope modified screening target in the world coordinate system can be proved.
In this paper, the imaging characteristics and motion characteristics of small targets in infrared images are analyzed from the perspective of space domain and time domain respectively, and a real-time infrared small target search and tracking algorithm based on adaptive track correlation is proposed. From the perspective of spatial domain, the curvature filtering algorithm is used to suppress the background according to the feature that the small target usually presents a sharp peak on the 3D surface with gray value as Z-axis. From the perspective of time domain, according to the difference of motion state between the target and the false alarm point, the track correlation method is used to further suppress the false alarm rate and realize the search and tracking of the target. In this paper, the concept of cosine similarity is introduced to judge the matching degree between suspected target and track on the basis of the traditional track correlation algorithm based on the idea of "nearest neighborhood", so as to reduce the probability of wrong track correlation. At the same time, based on residual analysis, the correlation gate size is selected adaptively and the observation results are smoothing by adaptive first-order low-pass filtering. Experiments show that the search and tracking algorithm proposed in this paper has real-time performance and high accuracy, and has good adaptability to different scenes.
Infrared (IR) small target detection in a single frame is a challenging task due to the lack of texture and color information and the interference of background clutters. In light of the two-dimensional Gaussian-like shape of IR small target, two properties from the perspective of local gradient and directional curvature (LGDC) are characterized. Specifically speaking, the local gradients in four quadrants as well as the curvatures from four directions should distribute in a regular way in the target region. Therefore, an LGDC map is computed from the input IR image so that the contrast between target and background can be greatly improved. By this means, we are able to extract the IR small target by a simple threshold related to the mean and standard deviation values of the LGDC map. Experiments implemented on real IR images verify that the proposed method can achieve satisfactory performance in terms of local contrast enhancement and background clutter suppression.
Research of modeling infrared radiation characteristic of ocean background plays an important role in areas like marine remote sensing, prevention and control of ocean pollution, meteorological observation, and so on. In this paper, we build a three-D ocean surface model based on P-M wave spectrum, then set up the camera projection model. We use LOWTRAN 7 to calculate solar irradiance and sky background radiance, and then use thermal radiation theory to calculate the thermal radiation of the ocean itself and use bidirectional reflectance distribution function to calculate the reflection of the sea to the solar radiation and sky background radiance. Finally, with all the above radiation components considered, we generate the ocean background infrared simulation image.
In this paper, we analyze the characteristics of pseudo-random code, by the case of m sequence. Depending on the description of coding theory, we introduce the jamming methods. We simulate the interference effect or probability model by the means of MATLAB to consolidate. In accordance with the length of decoding time the adversary spends, we find out the optimal formula and optimal coefficients based on machine learning, then we get the new optimal interference code. First, when it comes to the phase of recognition, this study judges the effect of interference by the way of simulating the length of time over the decoding period of laser seeker. Then, we use laser active deception jamming simulate interference process in the tracking phase in the next block. In this study we choose the method of laser active deception jamming. In order to improve the performance of the interference, this paper simulates the model by MATLAB software. We find out the least number of pulse intervals which must be received, then we can make the conclusion that the precise interval number of the laser pointer for m sequence encoding. In order to find the shortest space, we make the choice of the greatest common divisor method. Then, combining with the coding regularity that has been found before, we restore pulse interval of pseudo-random code, which has been already received. Finally, we can control the time period of laser interference, get the optimal interference code, and also increase the probability of interference as well.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.