Target Pose Measurement Based on Large Baseline Images

In order to address the problem of tracking different moving targets in image sequence against a complicated background, this paper presents a particle-filter multi-target tracking algorithm based on their dynamic salient features. By making use of the research findings on visual attention, the algorithm adopts the robust dynamic salient features as a result of combining the gray-scale and details with the motion characteristics of such targets as the state vector of particle filter. The algorithm is highly robust as it contains salient features originating from the low-level features of the targets. Meanwhile the particle filter allows optimized estimation of non-linear and non-Gaussian models. As a consequence, the algorithm is capable of managing traces in tracking different targets and dealing with their appearance, disappearance, mergence, splitting and sheltering by obstacles. Experiments show that this new algorithm enables tracking of multiple targets in complicated image sequence.

The conspicuity of different targets in image sequences taken by approaching sensors is addressed in applications such as the assessment of camouflage effectiveness or the performance evaluation of autonomous systems. In such evaluation processes the consideration of background characteristics is essential due to the propensity to confuse target and background signatures. Several discriminating features of target and background signature can be derived. Furthermore, the changing aspect and spatial resolution during an approach on a target have to be taken into account. Considering salient points in image sequences, we perform a nominal/actual value comparison by evaluating the receiver operating characteristic (ROC) curve for the detections in each image. Hence, reference regions for targets and backgrounds are provided for the entire image sequence by means of robust image registration. The consideration of the uncertainty for the temporal progression of the ROC curve enables hypothesis testing for well-founded statements about the significance of the determined distinctiveness of targets with respect to their background. The approach is neither restricted to images taken by IR sensors nor applicable to low level image analysis steps only, but can be considered as a general method for the assessment of feature evaluation and target distinctiveness. The analysis method proposed facilitates an objective comparison of object appearance with both, its relevant background and other targets, using different image analysis features. The feasibility and the usefulness of the approach are demonstrated with real data recorded with a FLIR sensor during a field trial on a bare and mock-up target.

This paper addresses the problem of designing an efficient and effective image sequence processing scheme that will successfully detect very small (point) targets in a cluttered background when both the target and clutter are moving through the image scene. The specific application area was detection of targets such as airplanes in infrared (IR) image sequences of a cloudy sky which have been taken by a stationary camera. In general we assume that targets are typically one to two pixels in extent and move only a fraction of a pixel per frame, are often low amplitude, and are found in scenes which also contain evolving clutter, e.g. clouds. Our algorithm is based on signal processing and detection theory, includes a perfect measurement performance analysis, and can be made computationally efficient compared to other approaches. Thus the algorithm could be applicable to other image sequence processing scenarios, using other acquisition systems besides IR, such as detection of small moving objects or structures in a biomedical or biological imaging scenario or the detection of satellites, meteors or other celestial bodies in night sky imagery acquired using a telescope. We present a GLRT solution, perfect measurement analysis including ROC curves, and results using real-world infrared data.

Face pose information is valuable for a variety of applications including unconstrained face recognition, natural human computer interfaces, and video database indexing. 3D pose estimation is a critical requirement for accurate face recognition using view varying representations, such as 2D intensity images. 3D pose extraction in this context requires 3D information, which is present in image sequences if we assume the moving objects in the sequence are primarily rigid. This paper presents a motion based pose estimation system which computes the 3D pose of the head in each frame of a video sequence. Generic low level features, such as corners, are identified and tracked in the video stream. The feature tracks are processed by a shape from motion algorithm which produces estimates of 3D geometry and pose. The geometry and pose estimates are considered together with facial structure constraints, temporal constraints, and initial pose estimates to refine knowledge of the specific face structure and its pose. We describe this system and show that it works on human faces. This is significant because the face has many smooth surfaces which make it difficult to extract dense intensity features.

This paper presents an approach for the detection of moving point targets in very low SNR based on purely time-domain analysis of pixels. The problem of detecting point targets in image sequences is formulated as a one-dimensional weak transient signal detection problem. The mathematical model of the transient signal is proposed. The key impact factors, such as the frame-rate, target velocity and SNR of the weak transient signal are studied. The detection ability is evaluated from the view of time-domain energy accumulation and correlation. The experimental results with both simulated and realworld data demonstrate that our approach can efficiently detect extremely low-SNR targets and the detection ability is decided by the key impact factors.

We describe an active contour based local energy minimization point distribution, behavior of orientation, relaxation iterative algorithm that estimates feature points of static target in image sequence. In our approach, we build a contour model of a target to get some of low-energy points kernels. The use of snake based line model results in more reliable convergence of the point local energy minimization. The algorithm uses auto-relation relation to give the behavior of orientation in a local kernel's window. It uses local window-based probability method to refine the current corresponding relations of scene kernels. Results are illustrated on real outdoor image sequence.

This paper describes a method to simultaneously estimate 3D pose and camera zoom parameters from sequential images. Given the polyhedral 3D model and its 2D surface texture, 3D pose parameters and camera focal lengths, which yield the best match between the current image and the reference image, are estimated precisely using gradient descent optimization. For the performance evaluation of the proposed algorithm, convergence tests were conducted. 3D pose and camera zoom tracking also conducted on both of synthesized and real sequential images. 3D object information makes the algorithm to effectively cope with self-occlusions, disappearance, and reappearance of partial surfaces of the object by checking visibility for each surface using its 3D pose. As the proposed method estimates camera focal lengths together with 3D rotation and translation, it can be applied to the 3D pose tracking on images of a camera with a zoom lens.

The relative pose (position and attitude) measurement of space noncooperative targets is very important for on-orbit servicing activities, such as target tracking, approaching, and capturing. The traditional methods rarely consider the instability of feature extraction and image blurring caused by target tumbling. In this paper, a method based on the maximum outer contour (MOC) recognition is proposed to measure the pose of the target. Different feature extraction algorithms can simultaneously achieve close- and long-range measurement tasks. First, the trailing image is restored by the image enhancement method. Second, the “rough + fine” combination recognition method is used for contour extraction and connected component labeling of the restored image, and the target feature extraction time is reduced to one-third of traditional methods. Furthermore, the elliptical surface on the MOC is fitted by the least squares method (LSM), and the ellipse parameters (i.e., the center position, the long- and short-axes, and the deflection angle) are extracted. The accuracy of the target recognition is improved. Third, for the close-range measurement, based on the detected ellipse parameter, the pose of the noncooperative target is solved by the binocular imaging algorithm of the space circle; for the long-range measurement, the contour centroid of the target is calculated by the detected MOC, and the position of the target is solved by the LSM. Moreover, the effectiveness of the method is verified by the OpenSceneGraph numerical simulation system. Finally, an experimental system consisting of a binocular camera, a UR5 manipulator, and a satellite mockup was built. The experimental results verified the proposed method.

Purpose. The aim of the article is to increase the efficiency of methods of spread of invasive plant species remote sensing to assess potential economic damage to the environment and agriculture through the use of intelligent surveillance systems. Methodology of research. The following methods were used in the study: monographic – for studying literature sources; machine learning method – for model training; three-way data split method for assessing model accuracy and statistical analysis for comparing recognition results; basic (traditional) methods for image enhancement; gradient image processing (Poisson mixing) – for image fusion. Findings. A model for improving the efficiency of methods for remote recognition of the spread of invasive plant species was developed and proposed. The performance of the proposed model was compared with the model based on traditional image augmentation methods. It was found that the CAIPIS method, which is the basis of the model, achieves significantly higher results in detecting invasive plant species, especially when the data set is limited. In addition, the method helps to increase the reliability and generalization of deep-learning models for detecting the spread of invasive plant species. The CAIPIS method provides means to create more diverse and realistic training samples, which helps to improve the generalization ability of the deep learning model in detecting invasive plants in aerial imagery of agricultural fields and ecosystems obtained using unmanned aerial vehicles. Originality. A new method for enhancing (augmenting) aerial images of crop fields is presented, which aims to improve the performance of invasive plant detection models by creating diverse and realistic training samples and increasing the variability of the training dataset to improve the performance of such models. Practical value. The application of the aerial images enhancement (augmentation) method will increase the effectiveness of controlling the spread of invasive pests in ecosystems. Key words: invasive species, phytoinvasion, augmentation, recognition, remote sensing.

This paper studies the problem of 3D reconstruction and pose estimation of non-cooperative targets in a space environment with uneven illumination. First, a simulation model of a failed satellite is constructed using SolidWorks to simulate the 3D structure and motion posture of a failed tumbling satellite. The satellite model is rendered with a single light source according to the characteristics of the space environment to construct a synthetic data set with good generalization ability. Then, in order to solve the problems of dark local details and blurred target information in the space environment, local enhancement is performed based on the LIME image enhancement algorithm, and foreground extraction is performed based on the Canny edge detection and morphological algorithm to improve the image quality and feature visibility. Based on the Colmap platform, incremental 3D reconstruction is performed through SIFT feature extraction and matching, triangulation, and bundle adjustment optimization to generate a sparse point cloud model and camera pose. On the basis of the reconstructed point cloud, the principal component analysis method is used to construct the target coordinate system, and the target pose is obtained through coordinate system transformation, which verifies the effectiveness of 3D reconstruction and pose estimation of non-cooperative targets in complex space environments.

Human pose estimation from video sequences has become a hot research topic in the domain of robotics and computer vision. However, existing three-dimensional (3D) pose estimation methods usually analyze individual frames, which have a low accuracy due to various human movement speed, limiting its practical application. In this paper, we propose a method for estimating 3D pose and calculating similarity from Tai Chi video sequences based on Seq2Seq network. Specifically, using 2D joint point coordinate sequence of the original image as input, our method constructs an encoder and a decoder to build a Seq2Seq network. Our method introduces an attention mechanism for weighing the input data to obtain an intermediate vector and decode it to estimate the 3D joint point sequence. Afterwards, using a template video and a target video as input, our method calculates the cost of passing through each point within the constraints to construct a cost matrix for video similarity. With the cost matrix, our method can determine the optimal path and use the correspondence of the video sequence to calculate the image similarity of the corresponding frame. The experimental data show that the proposed method can effectively improve the accuracy of 3D pose estimation, and increase the speed for video similarity calculation.

In this paper, we propose a novel hue-correction scheme based on the constant-hue plane in the RGB color space for deep-learning-based color-image enhancement. Existing hue-preserving image-enhancement methods cannot be applied to state-of-the-art enhancement methods such as deep-learning-based ones. Our main contributions are a discussion on the enhancement performance and the hue distortion of existing image-enhancement methods as well as the first attempt to make hue correction applicable to any existing image-enhancement methods including deep-learning-based ones. This novel scheme is carried out on the basis of the constant-hue plane in the RGB color space. In simulations, we first evaluated conventional image-enhancement methods in terms of the enhancement performance and hue distortion by using five objective metrics: the maximally saturated color similarity, the hue difference in CIEDE2000, discrete entropy, HIGRADE, and NIQMC. The experimental results show that recent deep-learning-based methods have a higher enhancement performance but cause images to be hue-distorted. In addition, the proposed scheme is demonstrated to be effective for suppressing hue distortion even under the use of deep-learning-based enhancement methods. Furthermore, it allows us not only to correct hue but also to maintain the performance of image-enhancement methods.

Computer vision researchers have proved the feasibility of camera self-calibration —the estimation of a camera's internal parameters from an image sequence without any known scene structure. Various self-calibration algorithms have been published. Nevertheless, all of the recent sequential approaches to 3D structure and motion estimation from image sequences which have arisen in robotics and aim at real-time operation (often classed as visual SLAM or visual odometry) have relied on pre-calibrated cameras and have not attempted online calibration.

Low-light images suffer from severe noise, low brightness, low contrast, etc. In previous researches, many image enhancement methods have been proposed, but few methods can deal with these problems simultaneously. In this paper, to solve these problems simultaneously, we propose a low-light image enhancement method that can be combined with supervised learning and previous HSV (Hue, Saturation, Value) or Retinex model-based image enhancement methods. First, we analyse the relationship between the HSV color space and the Retinex theory, and show that the V channel (V channel in HSV color space, equals the maximum channel in RGB color space) of the enhanced image can well represent the contrast and brightness enhancement process. Then, a data-driven conditional re-enhancement network (denoted as CRENet) is proposed. The network takes low-light images as input and the enhanced V channel (V channel of the enhanced image) as a condition during testing, and then it can re-enhance the contrast and brightness of the low-light image and at the same time reduce noise and color distortion. In addition, it takes 23 ms to process a color image with the resolution 400*600 on a 1080Ti GPU. Finally, some comparative experiments are implemented to prove the effectiveness of the method. The results show that the method proposed in this paper can significantly improve the quality of the enhanced image, and by combining it with other image contrast enhancement methods, the final enhancement result can even be better than the reference image in contrast and brightness when the contrast and brightness of the reference are not good.

Binocular camera systems are commonly used to construct 3-D-based scene description. However, there is a tradeoff between the length of the camera baseline and the difficulty of the matching problem and the extent of the field of view of the 3-D scene. A large baseline system provides better depth resolution than a smaller baseline system at the expense of a narrower field of view. To increase the depth resolution without increasing the difficulty of the matching problem and decreasing the field of view of the 3-D scene, a sequential 3-D-based scene description technique is proposed. Multiple small-baseline 3-D scene descriptions from a single moving camera or an array of cameras are used to sequentially construct a large baseline 3-D scene description while maintaining the field of view of a small-baseline system. A Bayesian framework using a disparity-space image (DSI) technique for disparity estimation is presented. The cost function for large baseline image matching is designed based not only on the photometric matching error, the smoothness constraint, and the ordering constraint, but also on the previous disparity estimates from smaller baseline stereo image pairs as a prior model. Texture information is registered along the scan path of the camera(s). Experimental results demonstrate the effectiveness of this technique in visual communication applications.