Line Segment Detector Research Articles

A Comprehensive Review of Image Line Segment Detection and Description: Taxonomies, Comparisons, and Challenges.

An image line segment is a fundamental low-level visual feature that delineates straight, slender, and uninterrupted portions of objects and scenarios within images. Detection and description of line segments lay the basis for numerous vision tasks. Although many studies have aimed to detect and describe line segments, a comprehensive review is lacking, obstructing their progress. This study fills the gap by comprehensively reviewing related studies on detecting and describing two-dimensional image line segments to provide researchers with an overall picture and deep understanding. Based on their mechanisms, two taxonomies for line segment detection and description are presented to introduce, analyze, and summarize these studies, facilitating researchers to learn about them quickly and extensively. The key issues, core ideas, advantages and disadvantages of existing methods, and their potential applications for each category are analyzed and summarized, including previously unknown findings. The challenges in existing methods and corresponding insights for potentially solving them are also provided to inspire researchers. In addition, some state-of-the-art line segment detection and description algorithms are evaluated without bias, and the evaluation code will be publicly available. The theoretical analysis, coupled with the experimental results, can guide researchers in selecting the best method for their intended vision applications. Finally, this study provides insights for potentially interesting future research directions to attract more attention from researchers to this field.

Tightly coupled stereo vision-inertial odometry based on point and line feature

Abstract To improve the accuracy and robustness of visual simultaneous localization and mapping (SLAM) in low-texture environments, this paper proposes a robust and fast stereo vision inertial SLAM pose estimation method that combines point and line features with an inertial measurement unit (IMU). The method tightly couples visual point and line features with IMU constraints, forming a least-squares problem through the error of each constraint term for nonlinear optimization. To address the issues of over-segmentation and time consumption in traditional line segment detection (LSD) algorithms, an improved LSD algorithm is adopted to accelerate line feature extraction. This approach merges nearby line segments based on spatial geometric relationships and filters out invalid segments, improving the time efficiency of the algorithm. Finally, experiments conducted in low-texture environments demonstrate that our algorithm achieves high localization accuracy and robustness.

A visual measurement method of vibration displacement of railway bridge bearings based on phase correlation

Currently, contact displacement meters for bearing displacement measurement are prone to fatigue failure. Non-contact visual measurement has become a more effective technique. In this study, an innovative model based on phase correlation is established for the visual measurement of vibration displacement of railway bridge bearings. Firstly, it is suggested to extract the vehicle–bridge interaction time using the sliding window average detection and K-means methods. Secondly, to circumvent the limited measurement accuracy of the existing methods, the phase correlation technique is applied to efficiently calculate the displacement of bearings within a two-dimensional plane. Thirdly, an image sharpness estimation method based on the Tenengrad function is proposed, and then a line segment detector is used for updating the measurement parameters. Through comparative experiments, the measurement error of presented method is within 0.04 mm. In the process of long-term service, the proposed method exhibits no fatigue failure issues, highlighting its good technical advantage.

Enhancing Deep Line Segment Detection and Performance Evaluation for Wood: A Deep Learning Approach with Experiment-Based, Domain-Specific Implementations

In recent decades, wood structures have gained significant attention for their ecological benefits and architectural versatility. The performance of wood, a popular construction material, often depends on the integrity of its connections. This study focuses on bolted glulam timber connections, which are strong but prone to cracks that pose structural health challenges. Traditional crack evaluation methods are manual, time-consuming, and error-prone. To address these issues, this research proposes a two-stage performance evaluation method. In the first stage, an innovative approach called ‘Enhanced Deep Line Segment Detection’ (Deep LSD), a non-supervised machine learning technique, is used for crack detection without relying on large, annotated datasets, thus enhancing efficiency and adaptability. In the second stage, cyclic loading assays simulate varying damage stages to collect data and establish a correlation between crack states and connection damage. The Park and Ang damage model is employed within this framework to assess the extent of damage. The efficacy of enhanced deep LSD is confirmed by comparing detected crack areas with ground truth measurements, yielding a high R-squared value of 0.98 and a minimal error margin of 1.41. Additionally, a damage index based on the Chinese standard (GB/T 24335-2009) is used to classify damage across different connection groups, ensuring robustness and alignment with established practices.

Run length encoding based weld seam detection from point clouds of ship stiffened panel

A vision-based weld seam detector is crucial for intelligent welding in ship stiffened panel as it influences the working efficiency of welding robots. This paper introduces an innovative Run Length Encoding (RLE) based Line Segment Detector (RSD) that is capable of extracting line segments from point clouds. The RLE is firstly employed to compress point cloud data by encoding them into runs. Then, the directional feature of significant line segments is determined according to the analysis of level line angles, facilitating a global line segment detection mechanism. This enables RSD to capture more complete line segments compared to region-based detection methods. Notably, this method obviates the need for manual parameter adjustments and does not require any prior information. The effectiveness and superiority of RSD are confirmed by testing various workpieces in a laboratory setting and comparing its performance with other detectors.

An integrated method for extracting channel network with check dams from high‐resolution DEM on the Loess Plateau

AbstractChannel networks have been widely used to model sediment transport and accumulation. Extracting channel networks in the check dam region from digital elevation models on the Loess Plateau can facilitate effective decision‐making and planning for soil and water conservation. Three methods are generally used to ensure the continuity of channel networks by removing check dams as hydrological barriers: filtering, filling and breaching. However, these methods may still cause disruption and displacement of the channel network owing to the existence of check dams. This study presents the development of an integrated method (improved regional growth and linear feature detection [iRG‐LFD]) for extracting natural continuous channel networks and locating check dams. First, a proposed improved region growth method based on channel and check dam terrain features was used to extract the complete channel network. Subsequently, the line segment detector for extracting straight lines was then improved to separate lines with different slopes. Finally, by combining the channel network and line segment detector results, a cross model was proposed for extracting check dams of different sizes. The experimental results for the Wangmaogou and Zhoutungou catchments showed minimal errors when the proposed method was used to extract the channel network, and F1‐scores of 86.67% and 86.95% were obtained for the predicted check dam samples in the two catchments, respectively. The results indicate that this method can be effectively used to extract continuous natural channel networks and accurately identify check dams and can thus be used to design soil and water conservation measures.

Line segment detectors with deformable attention

The object detectors based on Transformers are advancing rapidly. On the contrary, the development of line segment detectors is relatively slow. It is noteworthy that the object and line segments are both 2D targets. In this work, we design a line segment detection algorithm based on deformable attention. Leveraging this algorithm and the line segment loss function, we transform the object detectors, Deformable DETR and ViDT, into end-to-end line segment detectors named Deformable LETR and ViDTLE, respectively. In order to adapt the idea of sparse modeling for line segment detection, we propose a new attention mechanism named line segment deformable attention (LSDA). This mechanism focuses on the valuable positions under the guidance of reference line to refine line segments. We design an auxiliary algorithm named line segment iterative refinement for LSDA. With as few modifications as possible, we transform two object detection detectors, namely SMCA DETR and PnP DETR into competitive line segment detectors named SMCA LETR and PnP LETR, respectively. The experimental results show that the performances of the proposed methods are efficient.

Detecting Line Segments in Motion-Blurred Images With Events.

Making line segment detectors more reliable under motion blurs is one of the most important challenges for practical applications, such as visual SLAM and 3D line mapping. Existing line segment detection methods face severe performance degradation for accurately detecting and locating line segments when motion blur occurs. While event data shows strong complementary characteristics to images for minimal blur and edge awareness at high-temporal resolution, potentially beneficial for reliable line segment recognition. To robustly detect line segments over motion blurs, we propose to leverage the complementary information of images and events. Specifically, we first design a general frame-event feature fusion network to extract and fuse the detailed image textures and low-latency event edges, which consists of a channel-attention-based shallow fusion module and a self-attention-based dual hourglass module. We then utilize the state-of-the-art wireframe parsing networks to detect line segments on the fused feature map. Moreover, due to the lack of line segment detection datasets with pairwise motion-blurred images and events, we contribute two datasets, i.e., synthetic FE-Wireframe and realistic FE-Blurframe, for network training and evaluation. Extensive analyses on the component configurations demonstrate the design effectiveness of our fusion network. When compared to the state-of-the-arts, the proposed approach achieves the highest detection accuracy while maintaining comparable real-time performance. In addition to being robust to motion blur, our method also exhibits superior performance for line detection under high dynamic range scenes.

Digital quantification of the MMSE interlocking pentagon areas: a three-stage algorithm

The Mini-Mental State Examination (MMSE) is a widely employed screening tool for the severity of cognitive impairment. Among the MMSE items, the pentagon copying test (PCT) requires participants to accurately replicate a sample of two interlocking pentagons. While the PCT is traditionally scored on a binary scale, there have been limited developments of granular scoring scale to assess task performance. In this paper, we present a novel three-stage algorithm, called Quantification of Interlocking Pentagons (QIP) which quantifies PCT performance by computing the areas of individual pentagons and their intersection areas, and a balance ratio between the areas of the two individual pentagons. The three stages of the QIP algorithm include: (1) detection of line segments, (2) unraveling of the interlocking pentagons, and (3) quantification of areas. A set of 497 PCTs from 84 participants including their baseline and follow-up PCTs from the Rush Memory and Aging Project was selected blinded about their cognitive and clinical status. Analysis of the quantified data revealed a significant inverse relationship between age and balance ratio (beta = − 0.49, p = 0.0033), indicating that older age was associated with a smaller balance ratio. In addition, balance ratio was associated with perceptual speed (r = 0.71, p = 0.0135), vascular risk factors (beta = − 3.96, p = 0.0269), and medical conditions (beta = − 2.78, p = 0.0389). The QIP algorithm can serve as a useful tool for enhancing the scoring of performance in the PCT.

Yarn Angle Detection of Glass Fiber Plain Weave Fabric Based on Machine Vision

To address the issue of low accuracy in the yarn angle detection of glass fiber plain weave fabrics, which significantly impacts the quality and performance of the final products, a machine vision-based method for the yarn angle detection of glass fiber fabrics is proposed. The method involves pre-processing the image with brightness calculation, threshold segmentation, and skeleton extraction to identify the feature region. Line segment detection is then performed on this region, using the Hough transform. The concept of a “line segment evaluation index” is introduced, and it was used as a criterion for assessing the quality and relevance of detected line segments. Moreover, the warp and weft yarn extrusion area contours refer to the reconstructed outlines of yarn areas, achieved by combining the center of mass extraction with morphological operations and used to accurately determine the yarn angle. Tested under a range of challenging scenarios, including varied lighting conditions, fabric densities, and levels of image noise, this method has demonstrated robust stability and maintained high accuracy. These tests mimic real-world manufacturing environments, where factors such as ambient light changes and material inconsistencies can affect the quality of image capture and analysis. The proposed method has high accuracy, as shown by MSE and a Pearson’s r of 0.931. By successfully navigating these complexities, the proposed machine vision-based approach offers a significant enhancement in the precision of yarn angle detection for glass fiber fabric manufacturing, thus ensuring improved quality and performance of the final products.

ELSM: Evidence-Based Line Segment Merging

Abstract Existing line segment detectors break perceptually contiguous linear structures into multiple line segments. This can be offset by re-merging the segments, but existing merging algorithms over-merge and produce globally incorrect segments. Geometric cues are necessary but not sufficient for deciding whether to merge two segments or not. By restricting the result of any merging decision to have underlying image support, we reduce over-merging and globally incorrect segments. We propose a novel measure for evaluating merged segments based on line segment Hausdorff distance. On images from YorkUrbanDB, we show that our algorithm improves both qualitative and quantitative results obtained from four existing line segment detection methods and is better than two existing line segment merging methods. Our method does not suffer from inconsistent results produced by four recent deep learning-based models. The method is easily customisable to work for line drawings such as hand-drawn maps to obtain vectorized representations.

Review on line detection of wood panel images

In industrial, the modern intelligent degree of board counting is relatively low. Some small manufacturers still count by hand, but for large factories, manual counting needs a lot of human resources, and the accuracy of counting is low. With the rapid development of modern intelligence, image recognition is becoming more mature, some traditional algorithms keep emerging, such as Hough Transform, Fast Line Detector (FLD), Line Segment Detector (LSD) and other line detection algorithms, they have their own advantages and disadvantages, and are summarized and tested, compare which algorithm has higher accuracy and better effect in the field of board linear detection and counting. Finally, the operation mechanism, advantages and disadvantages are summarized, and the process and trend of the further optimization and development of the traditional algorithm line detection technology in the future are prospected, which provides some reference for the research in related fields.

Line segment detection algorithm in image extraction improvement study

In recent years, image processing technology has been developing and maturing, but due to the influence of many interfering factors in the acquisition process, there is a large amount of redundant information in the images obtained. The line segment detection algorithm in image extraction needs to be improved. This study utilizes computer technology to improve the line segment detection technology, and designs a line segment detection algorithm based on the linear detection improvement. Firstly, based on the basic principle of straight line detection algorithm, for the problems of line segment breakage and missing in straight line detection, RGB three-channel grayscale map is applied to detect line segments. Then the detected line segments are connected, merged and deleted. The test results show that the line segment detection algorithm improved based on straight line detection has the highest accuracy rate of 94.50 %, and the average processing time per image is also the lowest at 0.2 s. The algorithm runs faster at 0.25 s and has a higher F-value. It is able to detect the boundaries of a variety of rectangular targets, using the improved line segment detection algorithm has a wide range of applicability, lower error rate, and strong anti-interference ability. The improved line segment detection algorithm has a greater advantage in rectangular target extraction for document, text and book type images.

Line Segment Detection: a Review of the 2022 State of the Art

Line Segment Detection: a Review of the 2022 State of the Art

MPG-LSD: A high-quality line segment detector based on multi-scale perceptual grouping

Existing methods that rely on a single Gaussian scale to detect line segments could yield poor line continuity and inferior orientation and position accuracy due to insufficient suppression of quantization noise inherent in digital images. To address this fundamental issue, a novel multi-scale perceptual grouping-based line segment detector (MPG-LSD) is proposed in this paper. Our multi-scale perceptual grouping is developed to identify and aggregate collinear pixels that have similar gradient orientations for producing line segment candidates, not only over the input image but also across a set of Gaussian scales. Multi-scale line segment refinement and validation are then further developed and implemented to produce the final detection result, which delivers high quality in terms of line continuity, orientation and position accuracy. To enrich the evaluation of line segment detection performance, a new dataset consisting of high-resolution and natural noise-corrupted images with line segment annotations is constructed. Extensive experimental results show that our proposed MPG-LSD can outperform the current state-of-the-arts by a large margin.

Complex Background Removal Method for Video of an Inclined Cable Based on UAV Detection of Structural Vibration Frequencies

The safety of inclined cables is fundamental to the integrity of cable-stayed bridges. The vibrational frequencies of these cables form the foundation for assessing the cable force. Traditional contact measurement methods necessitate the installation of sensors on each cable, incurring substantial costs. In scenarios where camera placement adjacent to an inclined cable is impractical, noncontact approaches such as video capture via unmanned aerial vehicles prove effective. However, unmanned aerial vehicle-captured videos present a challenge due to their complex background, impeding cable feature recognition. In our study, we initially utilized the Region Growing algorithm for background subtraction. To enhance this method, we integrated it with the unique structural characteristics of cables, leading to the creation of the RGv2 algorithm. This novel algorithm offers increased processing speed and improved accuracy. Furthermore, we combined our method with empirical mode decomposition for effective detection of cable frequency characteristics. We also implemented a hybrid method, combining the K-Means and line segment detector algorithms with empirical mode decomposition. Compared to deep learning techniques for background subtraction, our proposed method demonstrates superior computational efficiency and promising potential for measuring vibrational frequencies of inclined cables.

A Fast Horizon Detector and a New Annotated Dataset for Maritime Video Processing

Accurate and fast sea horizon detection is vital for tasks in autonomous navigation and maritime security, such as video stabilization, target region reduction, precise tracking, and obstacle avoidance. This paper introduces a novel sea horizon detector from RGB videos, focusing on rapid and effective sea noise suppression while preserving weak horizon edges. Line fitting methods are subsequently employed on filtered edges for horizon detection. We address the filtering problem by extracting line segments with a very low edge threshold, ensuring the detection of line segments even in low-contrast horizon conditions. We show that horizon line segments have simple and relevant properties in RGB images, which we exploit to suppress noisy segments. Then we use the surviving segments to construct a filtered edge map and infer the horizon from the filtered edges. We propose a careful incorporation of temporal in- formation for horizon inference and experimentally show its effectiveness. We address the computational constraint by providing a vectorized implementation for efficient CPU execution, and leveraging image downsizing with minimal loss of accuracy on the original size. Moreover, we contribute a public horizon line dataset to enrich existing data resources. After extensive tests, we report the following major findings: 1) thanks to its filter, our algorithm accurately detects horizon lines with low or weak edge response, 2) the vectorized filter takes no more than 1.71% of the overall computations, while most of the computations are taken by the Line Segment Detection (LSD) algorithm we integrated into our pipeline, 3) our strategy of incorporating the temporal information avoids outlier detections, mitigates the effect of strong noisy lines, and exhibits high robustness when using incorrect detections as a temporal reference. Our algorithm’s performance is rigorously evaluated against state-of-the-art methods, and its core components are validated through ablation experiments.

DeeplabV3+-based navigation line extraction for the sunlight robust combine harvester.

Visual navigation is widely used in intelligent combine harvesters, but the existing algorithms do not have sufficiently high accuracy of the visual navigation line recognition under different sunlight conditions. To address this problem, this article proposes a sunlight-robust DeepLabV3+-based navigation line extraction method for combine harvesters. The navigation lines are extracted by constructing a new dataset and predicting the boundaries of the areas that have been and have not been cut. To address the problem that DeeplabV3+ is not sufficient light in the DCNN part, improvement is proposed by incorporating the MobileNetV2 module. In image segmentation, the prediction time is 22.5 ms, and the mean intersection over union (FMIOU) is 0.79. After image segmentation, the navigation lines are drawn using the line segment detection algorithm for the harvester. The proposed method is compared with other mainstream networks, and the prediction results are compared using the line segment detection method. The results show that this method can more quickly identify the navigation lines under different conditions of sunlight with less labeled data than the improved U-Net and DeeplabV3+, which uses Xception as the backbone. Compared to the traditional method and the improved U-Net, this method achieves good results and improves the recognition speed by 27 and 9 ms, respectively.

Line-Based 3D Building Abstraction and Polygonal Surface Reconstruction From Images.

Textureless objects, repetitive patterns and limited computational resources pose significant challenges to man-made structure reconstruction from images, because feature-points-based reconstruction methods usually fail due to the lack of distinct texture or ambiguous point matches. Meanwhile multi-view stereo approaches also suffer from high computational complexity. In this article, we present a new framework to reconstruct 3D surfaces for buildings from multi-view images by leveraging another fundamental geometric primitive: line segments. To this end, we first propose a new multi-resolution line segment detector to extract 2D line segments from each image. Then, we construct a 3D line cloud by introducing an improved Line3D++ algorithm to match 2D line segments from different images. Finally, we reconstruct a complete and manifold surface mesh from 3D line segments by formulating a Bayesian probabilistic modeling problem, which accurately generates a set of underlying planes. This output model is simple and has low performance requirements for hardware devices. Experimental results demonstrate the validity of the proposed approach and its ability to generate abstract and compact surface meshes from the 3D line cloud with low computational costs.

EvLSD-IED: Event-Based Line Segment Detection with Image-to-Event Distillation

EvLSD-IED: Event-Based Line Segment Detection with Image-to-Event Distillation