RESUMEN
Extracting circle information from images has always been a basic problem in computer vision. Common circle detection algorithms have some defects, such as poor noise resistance and slow computation speed. In this paper, we propose an anti-noise fast circle detection algorithm. In order to improve the anti-noise of the algorithm, we first perform curve thinning and connection on the image after edge extraction, then suppress noise interference by the irregularity of noise edges and extract circular arcs by directional filtering. In order to reduce the invalid fitting and speed up the running speed, we propose a circle fitting algorithm with five quadrants, and improve the efficiency of the algorithm by the idea of "divide and conquer". We compare the algorithm with RCD, CACD, WANG and AS on two open datasets. The results show that we have the best performance under noise while keeping the speed of the algorithm.
RESUMEN
Single-circle detection is vital in industrial automation, intelligent navigation, and structural health monitoring. In these fields, the circle is usually present in images with complex textures, multiple contours, and mass noise. However, commonly used circle-detection methods, including random sample consensus, random Hough transform, and the least squares method, lead to low detection accuracy, low efficiency, and poor stability in circle detection. To improve the accuracy, efficiency, and stability of circle detection, this paper proposes a single-circle detection algorithm by combining Canny edge detection, a clustering algorithm, and the improved least squares method. To verify the superiority of the algorithm, the performance of the algorithm is compared using the self-captured image samples and the GH dataset. The proposed algorithm detects the circle with an average error of two pixels and has a higher detection accuracy, efficiency, and stability than random sample consensus and random Hough transform.
RESUMEN
Circle detection is a fundamental problem in computer vision. However, conventional circle detection algorithms are usually time-consuming and sensitive to noise. In order to solve these shortcomings, we propose a fast circle detection algorithm based on information compression. First, we introduce the idea of information compression, which compresses the circular information on the image into a small number of points while removing some of the noise through sharpness estimation and orientation filtering. Then, the circle parameters stored in the information point are obtained by the average sampling algorithm with a time complexity of O(1) to obtain candidate circles. Finally, we set different constraints on the complete circle and the defective circle according to the sampling results and find the true circle from the candidate circles. The experimental results on the three datasets show that our method can compress the circular information in the image into 1% of the information points, and compared to RHT, RCD, Jiang, Wang and CACD, Precision, Recall, Time and F-measure are greatly improved.
RESUMEN
RGB-Depth (RGB-D) cameras are widely used in computer vision and robotics applications such as 3D modeling and humanâ»computer interaction. To capture 3D information of an object from different viewpoints simultaneously, we need to use multiple RGB-D cameras. To minimize costs, the cameras are often sparsely distributed without shared scene features. Due to the advantage of being visible from different viewpoints, spherical objects have been used for extrinsic calibration of widely-separated cameras. Assuming that the projected shape of the spherical object is circular, this paper presents a multi-cue-based method for detecting circular regions in a single color image. Experimental comparisons with existing methods show that our proposed method accurately detects spherical objects with cluttered backgrounds under different illumination conditions. The circle detection method is then applied to extrinsic calibration of multiple RGB-D cameras, for which we propose to use robust cost functions to reduce errors due to misdetected sphere centers. Through experiments, we show that the proposed method provides accurate calibration results in the presence of outliers and performs better than a least-squares-based method.
RESUMEN
Given a spherical set of points, we consider the detection of cocircular subsets of the data. We distinguish great circles from small circles, and develop algorithms for detecting cocircularities of both types. The suggested approach is an extension of the Hough transform. We address the unique parameter-space quantization issues arising due to the spherical geometry, present quantization schemes, and evaluate the quantization-induced errors. We demonstrate the proposed algorithms by detecting cocircular cities and airports on Earth's spherical surface. These results facilitate the detection of great and small circles in spherical images.
RESUMEN
In this paper, a stereo vision 3D position measurement system for a three-axial pneumatic parallel mechanism robot arm is presented. The stereo vision 3D position measurement system aims to measure the 3D trajectories of the end-effector of the robot arm. To track the end-effector of the robot arm, the circle detection algorithm is used to detect the desired target and the SAD algorithm is used to track the moving target and to search the corresponding target location along the conjugate epipolar line in the stereo pair. After camera calibration, both intrinsic and extrinsic parameters of the stereo rig can be obtained, so images can be rectified according to the camera parameters. Thus, through the epipolar rectification, the stereo matching process is reduced to a horizontal search along the conjugate epipolar line. Finally, 3D trajectories of the end-effector are computed by stereo triangulation. The experimental results show that the stereo vision 3D position measurement system proposed in this paper can successfully track and measure the fifth-order polynomial trajectory and sinusoidal trajectory of the end-effector of the three- axial pneumatic parallel mechanism robot arm.
Asunto(s)
Percepción de Profundidad , Imagenología Tridimensional/métodos , Presión , Robótica/instrumentación , Robótica/métodos , Algoritmos , Calibración , Modelos LinealesRESUMEN
In this paper, we propose a new approach to detect circles and nano-particles based on an oriented-edges gradient map and a decision tree. The decision tree is calculated from geometric constraints based on particular right triangles inscribed in a circle. Use of the proposed accumulator and dynamic storage matrix radii shows the robustness of our algorithm in terms of results and execution time. This robustness can also be enhanced in the event of prior knowledge. Indeed, we can enable or disable intermediate nodes or a part of nodes of the proposed decision tree to strengthen both the detection results and the execution time of the algorithm. Our approach makes it possible to detect circles and analyse the distribution of the nano-particles which is evaluated using four databases which include TEM, synthetic, real and complex images.
RESUMEN
The contour of an arbitrary figure can be represented as a group of circles of curvature in contact with it, with each curvature circle represented by its center OC and radius r. We propose a series of cell models for detecting this circle, which is composed of a lateral geniculate nucleus (LGN) cell, nondirectionally selective (NDS) simple cell, and curvature-circle detection cell (CDC). The LGN and NDS simple cells were previously modeled. The CDC has been modeled as follows. Each tangent in contact with this circle is detected by an NDS simple cell that performs the Hough transformation of LGN cell responses, and then this tangent is transformed to a three-dimensional (3D) normal line in a CDC column. This transformation has been named a 3D normal-line transform. Performing this transformation for all tangents causes a CDC at the intersection of these normal lines to fire most intensively, and thus the OC and r of the circle is detected as the coordinates of this intersection. Therefore, the CDC has been modeled as this 3D normal-line transform. Based on this CDC, we model two types of constancy CDC: a position-invariant CDC and a curvature-invariant CDC. These three types of CDC reflect the response to various stimuli in actual area V4 cells. In order to validate these CDC types neurophysiologically, we propose an experimental method using microelectrodes. Cell models previously reported correspond to this hierarchy: the S1, S2, and C2 cells correspond to the NDS simple cell, CDC, and position-invariant CDC, respectively.
RESUMEN
The Hough transform has been widely used in image analysis and digital image processing due to its capability of transforming image space detection to parameter space accumulation. In this paper, we propose a novel Angle-Aided Circle Detection (AACD) algorithm based on the randomized Hough transform to reduce the computational complexity of the traditional Randomized Hough transform. The algorithm ameliorates the sampling method of random sampling points to reduce the invalid accumulation by using region proposals method, and thus significantly reduces the amount of computation. Compared with the traditional Hough transform, the proposed algorithm is robust and suitable for multiple circles detection under complex conditions with strong anti-interference capacity. Moreover, the algorithm has been successfully applied to the welding spot detection on automobile body, and the experimental results verifies the validity and accuracy of the algorithm.
Asunto(s)
Automóviles , Procesamiento de Imagen Asistido por Computador/métodos , Aprendizaje Automático , Reconocimiento de Normas Patrones Automatizadas/métodos , Soldadura , Algoritmos , Automatización , Análisis por Conglomerados , Diseño de Equipo , Programas InformáticosRESUMEN
Objective To compare the principles and performance of three Hough transform algorithms(standard Hough transform,gradient based Hough transform,and random Hough transform)in order to establish a suitable control basis for precise and rapid recognition of targets and acquisition of target center coordinates for craniocerebral puncture robots.Methods A simulation environment in MATLAB software was built to study and analyze image feature recognition,filtering,edge detection,cumulative voting and other processing engineering.Contour recognition and fitting of target circles were achieved in multiple scenarios before their center coordinates were obtained.The recognition and fitting performance of these algorithms was quantitatively compared.Finally,a better detection algorithm based on the actual environment of the craniocerebral puncture robot was determined.Results The standard Hough transform algorithm had the largest error between the mark circle and the target circle,and the running time of this algorithm was the longest due to large computation.The detection speed of the random Hough transform algorithm was lower than that of the gradient-based Hough transform algorithm,but the fitting accuracy was slightly better than that of the standard Hough transform algorithm.The speed and accuracy of circle fitting based on the gradient Hough transform algorithm had significant advantages over the other two.Conclusion The gradient based Hough transform algorithm is more suitable for obtaining the target center coordinates of the craniocerebral puncture robot system.