Line Segment Detector Research Articles

Automatic comic page segmentation based on polygon detection

Comic page segmentation aims to automatically decompose scanned comic images into storyboards (frames), which is the key technique to produce digital comic documents that are suitable for reading on mobile devices. In this paper, we propose a novel method for comic page segmentation by finding the quadrilateral enclosing box of each storyboard. We first acquire the edge image of the input comic image, and then extract line segments with a heuristic line segment detection algorithm. We perform line clustering to further merge the overlapped line segments and remove the redundancy line segments. Finally, we perform another round of line clustering and post-processing to compose the obtained line segments into complete quadrilateral enclosing boxes of the storyboards. The proposed method is tested on 2,237 comic images from 12 different printed comic series, and the experimental results demonstrate that our method is effective for comic image segmentation and outperforms the existing methods.

AUTOMATIC PROCEDURE FOR THE REGISTRATION OF THERMOGRAPHIC IMAGES WITH POINT CLOUDS

Abstract. This paper presents a procedure for the automatic registration of thermographies with laser scanning point clouds. Given the heterogeneous nature of the two modalities, we propose a feature-based approach, satisfying the requisite that extracted features have to be invariant not only to rotation, translation and scale but also to changes in illumination and dimensionality. As speed and minimum operator interaction are prerequisites for the viability of the process in the building industry, our automatic registration procedure includes automatic feature extraction with no human intervention. With this aim, a line segment detector is used to extract 2D lines from thermographies, and 3D lines are extracted through segmentation of the point cloud. Feature-matching and the relative pose between thermographies and point cloud are obtained from an iterative procedure applied to detect and reject outliers; this includes rotation matrix and translation vector calculation and the application of the RANSAC algorithm to find a consistent set of matches. An automatically textured thermographic 3D model is the expected result of these procedures once the point cloud is filtered and triangulated.

Beyond Independence: An Extension of the A Contrario Decision Procedure

The a contrario approach is a principled method for making algorithmic decisions that has been applied successfully to many tasks in image analysis. The method is based on a background model (or null hypothesis) for the image. This model relies on independence assumptions and characterizes images in which no detection should be made. It is often image dependent, relying on statistics gathered from the image, and therefore adaptive. In this paper we propose a generalization for background models which relaxes the independence assumption and instead uses image dependent second order properties. The second order properties are accounted for thanks to graphical models. The modified a contrario technique is applied to two tasks: line segment detection and part-based object detection, and its advantages are demonstrated. In particular, we show that the proposed method enables reasonably accurate prediction of the false detection rate with no need for training data.

A monocular camera gyroscope

We present a method for tracking the 3-axis orientation of a monocular camera using orthogonal vanishing points detected in individual frames of a sequence of images. Robust and real-time vanishing point detection is done using a standard line segment detection method and an adaptive RANSAC algorithm. Vanishing points and corresponding vanishing directions found in consecutive frames are associated with each other to produce a sequence of orientation quaternions, which is processed by an extended Kalman filter. Experiments with a consumer-level, handheld mobile device indicate that the accuracy of the proposed method is comparable with those of consumer-grade mechanical motion sensors.

LSD: a Line Segment Detector

LSD is a linear-time Line Segment Detector giving subpixel accurate results. It is designed to work on any digital image without parameter tuning. It controls its own number of false detections: On average, one false alarms is allowed per image. The method is based on Burns, Hanson, and Riseman's method, and uses an a-contrario validation approach according to Desolneux, Moisan, and Morel's theory. The version described here includes some further improvement over the one described in the original article.

一种基于梯度的直线段检测算法

An algorithm for line segments detection in digital image was proposed to improve the computation and the robustness of the traditional algorithm.The image gradient is meaningful to detect the edge of structure in the image;therefore its module and direction are computed in the first step.According to the pseudo-sort result of the gradient module,the region-growing method is implemented to enlarge the pixels with the same direction,where the connected pixel region is considered as a candidate region of line segments.Finally,enclosing rectangle is used to describe and judge the candidate region by long axis and short axis of the rectangle,belong which the long axis that meets the requirement is the desired line segment.Based on MATLAB program,an experimental simulation was performed.Experimental results depicted that 108 line segments after 8.87s of computation were detected by the proposed algorithm;which saved 17% of computation time and also detected 16% line segments more than the traditional LSD algorithm.This algorithm can accurately and quickly detect line segments in many complex environments in the image.

Robust integral image rectification framework using perspective transformation supported by statistical line segment clustering

In most integral image analysis and processing tasks, accurate knowledge of the internal image structure is required. In this paper we present a robust framework for the accurate rectification of perspectively distorted integral images based on multiple line segment detection. The use of multiple line segments increases the overall fault tolerance of our framework providing strong statistical support for the rectification process. The proposed framework is used for the automatic rectification, metric correction, and rotation of distorted integral images. The performance of our framework is assessed over a number of integral images with varying scene complexity and noise levels.

Road Detection and Tracking from Aerial Desert Imagery

We present a fast, robust road detection and tracking algorithm for aerial images taken from an Unmanned Aerial Vehicle. A histogram-based adaptive threshold algorithm is used to detect possible road regions in an image. A probabilistic hough transform based line segment detection combined with a clustering method is implemented to further extract the road. The proposed algorithm has been extensively tested on desert images obtained using an Unmanned Aerial Vehicle. Our results indicate that we are able to successfully and accurately detect roads in 96% of the images. We experimentally validated our algorithm on over a thousand aerial images obtained using our UAV. These images consist of straight and curved roads in various conditions with significant changes in lighting and intensity. We have also developed a road-tracking algorithm that searches a local rectangular area in successive images. Initial results are presented that shows the efficacy and the robustness of this algorithm. Using this road tracking algorithm we are able to further improve the road detection and achieve a 98% accuracy.

Biologically inspired EM image alignment and neural reconstruction

Three-dimensional reconstruction of consecutive serial-section transmission electron microscopy (ssTEM) images of neural tissue currently requires many hours of manual tracing and annotation. Several computational techniques have already been applied to ssTEM images to facilitate 3D reconstruction and ease this burden. Here, we present an alternative computational approach for ssTEM image analysis. We have used biologically inspired receptive fields as a basis for a ridge detection algorithm to identify cell membranes, synaptic contacts and mitochondria. Detected line segments are used to improve alignment between consecutive images and we have joined small segments of membrane into cell surfaces using a dynamic programming algorithm similar to the Needleman-Wunsch and Smith-Waterman DNA sequence alignment procedures. A shortest path-based approach has been used to close edges and achieve image segmentation. Partial reconstructions were automatically generated and used as a basis for semi-automatic reconstruction of neural tissue. The accuracy of partial reconstructions was evaluated and 96% of membrane could be identified at the cost of 13% false positive detections. An open-source reference implementation is available in the Supplementary information. seymour.kb@ed.ac.uk; douglas.armstrong@ed.ac.uk Supplementary data are available at Bioinformatics online.

EDLines: A real-time line segment detector with a false detection control

We propose a linear time line segment detector that gives accurate results, requires no parameter tuning, and runs up to 11 times faster than the fastest known line segment detector in the literature; namely, the line segment detector (LSD) by Grompone von Gioi et al. The proposed algorithm makes use of the clean, contiguous (connected) chain of edge pixels produced by our novel edge detector, the Edge Drawing ( ED) algorithm; hence the name EDLines. The detector includes a line validation step due to the Helmholtz principle, which lets it control the number of false detections. With its accurate results and blazing speed, EDLines will be very suitable for the next generation real-time computer vision and image processing applications.

A novel Hough transform method for line detection by enhancing accumulator array

In this paper, an improved Hough transform (HT) method is proposed to robustly detect line segments in images with complicated backgrounds. The work focuses on detecting line segments of distinct lengths, totally independent of prior knowledge of the original image. Based on the characteristics of accumulation distribution obtained by conventional HT, a local operator is implemented to enhance the difference between the accumulation peaks caused by line segments and noise. Through analysis of the effect of the operator, a global threshold is obtained in the histogram of the enhanced accumulator to detect peaks. Experimental results are provided to demonstrate the efficiency and robustness of the proposed method.

Line segment detection using weighted mean shift procedures on a 2D slice sampling strategy

A new line segment detection approach is introduced in this paper for its application in real-time computer vision systems. It has been designed to work unsupervised without any prior knowledge of the imaged scene; hence, it does not require tuning of input parameters. Although many works have been presented on this topic, as far as we know, none of them achieves a trade-off between accuracy and speed as our strategy does. The reduction of the computational cost compared to other fast methods is based on a very efficient sampling strategy that sequentially proposes points on the image that likely belong to line segments. Then, a fast line growing algorithm is applied based on the Bresenham algorithm, which is combined with a modified version of the mean shift algorithm to provide accurate line segments while being robust against noise. The performance of this strategy is tested for a wide variety of images, comparing its results with popular state-of-the-art line segment detection methods. The results show that our proposal outperforms these works considering simultaneously accuracy in the results and processing speed.

An Approach for Detecting Regular Hexagon

Regular hexagon is a kind of common geometric figure. This paper proposes a vision detection method based on geometric characteristics to detect regular hexagon. First, line segments are detected in the image by hough transform. The line equations are constructed according to the detected line segments in the image space. Second, the line equations are utilized to compute the intersection points of adjacent two lines. The lengths of every side of regular hexagon are gotten by computing the distances between adjacent two intersection points. The angles between adjacent lines are calculated. Then the circle detection method based hough transform is employed to inspect whether all intersection points locate at the same circumference. Finally, the detection experiment of hexagon bolt is constructed to illustrate the performance of the presented method. The measurement errors of angles between adjacent two sides and lengths of sides of regular hexagon are 0.38% and 0.85% respectively. The result shows that the presented method is effective and can precisely detect regular hexagon in the image.

Automatic road extraction in rural areas, based on the Radon transform using digital images

Manual digitizing of road maps is an expensive and time-consuming task. We present a method based on the Radon transform, which automatically extracts roads in rural areas from digital images. The proposed method iteratively detects linear segments and then generates road centerlines. The proposed method is composed of two phases: in the first, the seed segments are detected, and in the second, road widths are measured and successive detection of line segments that are candidates as road centerlines is performed. Several tests were carried out using aerial photographs (digital images) of a rural area in Brazil, and the results are presented and discussed. The quality of the extracted road centerlines was computed using the following indexes: completeness, correctness, and RMS. The values obtained show that the proposed methodology performs well.

LSD: A Fast Line Segment Detector with a False Detection Control

We propose a linear-time line segment detector that gives accurate results, a controlled number of false detections, and requires no parameter tuning. This algorithm is tested and compared to state-of-the-art algorithms on a wide set of natural images.

Optimization models and algorithms for the hyperplane clustering problem

This is a summary of the author’s PhD thesis supervised by Edoardo Amaldi and defended on 3 April 2009 at the Politecnico di Milano. The thesis is written in English and is available from the author upon request. In this work, we extensively study two challenging variants of the general problem of clustering a given set of data points with respect to hyperplanes so as to extract collinearity between them. After pointing out the similarities and differences with previous work on related problems, we propose mathematical programming formulations for our problem variants. Since these problems are difficult to handle due to the nonlinear nonconvexity that arises because of the l2-norm in the distance function and a large number of binary assignment variables, we develop column generation algorithms and heuristics to tackle them. The efficiency of the methods developed is demonstrated on realistic randomly generated instances along with applications in piecewise linear model fitting and line segment detection in digital images.

A New Method on Matching Correspondence Features in Image Warping

This paper introduces a robust method for automatically matching features in images corresponding to the same physical point on an object captured from two arbitrary viewpoints. Starting from detected line segments in two or more images using phase congruency, the pairs of line segments are characterized by indices that encode the relative positions and orientations of those segments. The encoding indices are invariant with respect to viewpoint changes and the corresponding line segments between the images can be found by them. The feature matching is optimized for image warping where we wish to ignore unreliable matches at the expense of reducing the number of feature matches. This approach can be applied to the image warping to realize its automatism. Results are presented on real images.

Real-time line segments detection based on graphic processor

An iterative line segment detection procedure by graphic hardware was presented,which implemented edge detection,Hough transform,end points detection with GPU.The GPU based parallel Hough transform was designed for both θ parameter space segmentation and image space segmentation,and threads for the computation of the same θ parameter space were properly synchronized by shared memory in thread block.Experimental results prove the accuracy of the method for its iterative property,and good real-time capability because of full utilization of GPU's powerful computation effort.

Automated Geospatial Conflation of Vector Road Maps to High Resolution Imagery

As the availability of various geospatial data increases, there is an urgent need to integrate multiple datasets to improve spatial analysis. However, since these datasets often originate from different sources and vary in spatial accuracy, they often do not match well to each other. In addition, the spatial discrepancy is often nonsystematic such that a simple global transformation will not solve the problem. Manual correction is labor-intensive and time-consuming and often not practical. In this paper, we present an innovative solution for a vector-to-imagery conflation problem by integrating several vector-based and image-based algorithms. We only extract the different types of road intersections and terminations from imagery based on spatial contextual measures. We eliminate the process of line segment detection which is often troublesome. The vector road intersections are matched to these detected points by a relaxation labeling algorithm. The matched point pairs are then used as control points to perform a piecewise rubber-sheeting transformation. With the end points of each road segment in correct positions, a modified snake algorithm maneuvers intermediate vector road vertices toward a candidate road image. Finally a refinement algorithm moves the points to center each road and obtain better cartographic quality. To test the efficacy of the automated conflation algorithm, we used U.S. Census Bureau's TIGER vector road data and U.S. Department of Agriculture's 1-m multi-spectral near infrared aerial photography in our study. Experiments were conducted over a variety of rural, suburban, and urban environments. The results demonstrated excellent performance. The average correctness measure increased from 20.6% to 95.5% and the average root-mean-square error decreased from 51.2 to 3.4 m.

Line detection based on Euclidean distance

To detect lines in digital image, this paper proposed a new algorithm. Different form traditional line detection algorithms, the algorithm does not search another coordinate transformation, but defines the line degree of a point in a curve as the sum of k Euclidean distance square between points centered at the point, and deduces that except limited points in the head and tail, points in a line segment will have the maximum line degree. Derived from this characteristic, the new algorithm detects line segments in a two-pass manner. Firstly, the points on a curve that obviously belonged to a line segment were selected by using Freeman chain-code. Secondly, the points that were continuous and had the maximum line degree were selected. The result shows the algorithm is stable, precise and fast.