Adjacent Feature Points Research Articles

Dimensional synthesis of motion generation in a spherical four-bar mechanism

A new objective function for optimizing spherical, four-bar mechanisms for motion generation is presented. Rigid-body poses of a spherical four-bar mechanism in a standard installation position are investigated, and a normalization processing method is proposed. After processing the coupler points of the mechanism, the feature points generated are located on the circles. The formation principle of the feature coupler circles is analysed, and the internal relationship between the centre angle of the adjacent feature points on the feature coupler circles and the coupler angle of the spherical four-bar mechanism is determined. An input angle determination method is proposed for the spherical four-bar mechanism in a general installation position. An objective function is then established to optimize the basic dimensional types, relative input angles, and installation position parameters of the desired spherical four-bar mechanism. The proposed method is applicable to both prescribed and unprescribed timing problems. Notably, the dimension of the optimization variables is only eight, which is exceptionally low for multiple position motion generation without prescribed timing; therefore, the global optimal solution is more easily obtained. The optimization process is carried out by the genetic algorithm. The feasibility and effectiveness of our proposed method are demonstrated by examples.

Navigation Line Extraction Method for Broad-Leaved Plants in the Multi-Period Environments of the High-Ridge Cultivation Mode

Navigation line extraction is critical for precision agriculture and automatic navigation. A novel method for extracting navigation lines based on machine vision is proposed herein using a straight line detected based on a high-ridge crop row. Aiming at the low-level automation of machines in field environments of a high-ridge cultivation mode for broad-leaved plants, a navigation line extraction method suitable for multiple periods and with high timeliness is designed. The method comprises four sequentially linked phases: image segmentation, feature point extraction, navigation line calculation, and dynamic segmentation horizontal strip number feedback. The a* component of the CIE-Lab colour space is extracted to preliminarily extract the crop row features. The OTSU algorithm is combined with morphological processing to completely separate the crop rows and backgrounds. The crop row feature points are extracted using an improved isometric segmented vertical projection method. While calculating the navigation lines, an adaptive clustering method is used to cluster the adjacent feature points. A dynamic segmentation point clustering method is used to determine the final clustering feature point sets, and the feature point sets are optimised using lateral distance and point line distance methods. In the optimisation process, a linear regression method based on the Huber loss function is used to fit the optimised feature point set to obtain the crop row centreline, and the navigation line is calculated according to the two crop lines. Finally, before entering the next frame processing process, a feedback mechanism to calculate a number of horizontal strips for the next frame is introduced to improve the ability of the algorithm to adapt to multiple periods. The experimental results show that the proposed method can meet the efficiency requirements for visual navigation. The average time for the image processing of four samples is 38.53 ms. Compared with the least squares method, the proposed method can adapt to a longer growth period of crops.

Deep learning‐based super‐resolution with feature coordinators preservation for vision‐based measurement

Vision-based displacement measurement is promising and emerging for structural monitoring. However, the accuracy of visual measurement is commonly limited by the resolution of the camera. The super-resolution (SR) technique can reconstruct high-resolution images from the corresponding low-resolution images within the constraints of prior knowledge. Existing SR algorithms mainly focus on improving the overall quality of the image. By contrast, the accurate extraction of the coordinates of feature points is the most important for the visual measurement. Besides, the SR network is usually trained by an artificial dataset whose low-resolution images are obtained by artificially degrading the corresponding high-resolution images, instead of those directly captured by cameras. However, this degradation usually is only a simple bicubic downsampling that cannot reflect the real degradation, which will provide inaccurate constraints to the network training. Therefore, this paper proposes a novel SR framework that can significantly preserve the feature coordinators for visual measurement (SRFCP). First, a deep learning-based SR network that focuses on feature preservation is proposed, which introduces both feature weighted branch and feature preserving loss. Second, an image degradation model is built based on the blur kernel and noise extracted from the images captured in real scene. Experiments on public datasets show that the proposed SRFCP performs well both in terms of the objective evaluation index and the subjective visual effect. Then, a binocular visual measurement platform is set up to measure the distance of adjacent feature points on a chessboard. Lastly, several SR algorithms are evaluated by the improvement they bring to the measurement accuracy. Experimental results show that the distance measurement performance can be significantly improved by the images reconstructed by the SRFCP. Therefore, the proposed SRFCP can accurately reconstruct the high-resolution images while preserving the features coordinates, which is crucial for the visual measurement in structural monitoring.

Network Communication Platform Based on Artificial Intelligence

In order to improve the integration of digital prints, this paper proposes a new method of printmaking analog synthesis, that is, using interactive technology.. Firstly, Harris corner algorithm is used to collect and preprocess the adjacent feature points of digital printmaking image, and the texture features of digital printmaking image are extracted, so as to construct the texture information transmission model of digital printmaking image; Secondly, with the help of segmentation technology to process the digital print image, the dynamic feature segmentation is carried out, the local binary fitting method is used to enhance and repair the digital print image information, and the information fusion method based on interactive technology is used to complete the analog synthesis of digital print image; Finally, the simulation results show that the method has good performance of simulation synthesis and strong information fusion ability.

Lifting Scheme-Based Sparse Density Feature Extraction for Remote Sensing Target Detection

The design of backbones is of great significance for enhancing the location and classification precision in the remote sensing target detection task. Recently, various approaches have been proposed on altering the feature extraction density in the backbones to enlarge the receptive field, make features prominent, and reduce computational complexity, such as dilated convolution and deformable convolution. Among them, one of the most widely used methods is strided convolution, but it loses the information about adjacent feature points which leads to the omission of some useful features and the decrease of detection precision. This paper proposes a novel sparse density feature extraction method based on the relationship between the lifting scheme and convolution, which improves the detection precision while keeping the computational complexity almost the same as the strided convolution. Experimental results on remote sensing target detection indicate that our proposed method improves both detection performance and network efficiency.

RETRACTED: Online position recognition and correction method for sports athletes

Abstract In swimming, the correct recognition and correction of the wrong posture of the swimmer can improve the training quality of the athletes on weekdays. When the posture recognition is corrected, the affine deformation of the human body is easy to occur in the course of swimming. The traditional method is to extract these feature points and compare them with the correct posture to realize the recognition and correction of the posture. Due to the failure to detect and correct the wrong posture of athletes in real time, a method of position recognition and correction for swimmers with depth image bone tracking is proposed, and the threshold method is used to preprocess the image. The kalian filter is used to filter the collected image, Ago Si distribution function is used to obtain the action feature points from the filtered image, and the marginal point and the action feature point with low brightness are screened out by surf method. The Euclidean distance method is used to determine the distance between two adjacent feature points, and the feedback monitoring principle is used to identify and correct the wrong posture. The simulation results show that the improved pose recognition and correction method for skeletal tracking in depth image is improved. It can track and monitor the movement of athletes and complete the detection and recognition of swimming posture with high accuracy and strong stability.

A new calibrator providing easy detection of feature points for calibrating fisheye cameras in vehicle AVM systems

This paper proposes a new planar calibrator, which is suitable for calibrating the fisheye camera in a vehicle around view monitoring system. To facilitate easy feature point detection using the proposed calibrator, we ensure that the shapes of the features and the distances between the adjacent feature points on the calibrator’s image acquired by the fisheye camera are identical squares and equal, respectively, unlike those on the conventional chessboard calibrator. Further, the optimum number and locations of feature points on the proposed calibrator are experimentally determined for ensuring the best camera calibration performance. Compared with the conventional chessboard calibrator and the H-pattern calibrator, the proposed calibrator shows superior performance in terms of feature point detection and camera calibration.

Mine Laneway Modeling and Realization Based on the Matlab

In view of the request of 3D visualization on mine laneway, according to traverse measurement data and profile data, by use of triangulated irregular network, the shortest dagonal method and the adjacent feature points to pace method ,using Matlab language, the 3D laneway body were established. The result shows that the method is feasible and has a simple and fast reconstruction method.

Match Planar Curve Based on Affine Invariant

In order to recognize the curve with the same feature points but the curvature between the two adjacent feature points is different; a novel method for matching curve is proposed under the affine transformation. First feature points are matched both on model contour and object contour, then the affine region is divided for obtaining a series of sub-region's centroids by using feature points and the whole contour's centroid. Because area ratio of triangles, which constituted by centroid of the whole contour region and sub-region’s centroids, is invariant under affine transformation. We introduce area ratio of triangles to construct affine invariant, and normalized it too. And then a new recognition vector matrix is defined, and a different measure function is defined by us too. Finally the different measure value between model curve and object curve is calculated, which can be employed to judge whether the model and object are matched. Theory analysis and experiment results show the method is feasible, which can not only match the different curve but also distinguish the object with the similar shape according to the different measure value.

Vehicle Model Recognition Based on Fuzzy Pattern Recognition Method

The method based on the theory of Fuzzy Pattern Recognition is divided into three parts. Firstly, use Hough transformation to extract the feature points of vehicles, and use the ratio between two absolute distance of adjacent feature points as the characteristic values of vehicles; secondly, use Fuzzy C-mean Classification to handle feature data of 75 car model, then establish a degree of membership matrix as the sample space; thirdly, consider the classification algorithm based on fuzzy approach degree and the credibility of the vehicle feature to propose a weighted close- degree recognition algorithm. This recognition method has a good effect.