JP2023120281A - System and method for detecting line in vision system - Google Patents

Abstract

To provide a system and a method for detecting a plurality of line features within an image.SOLUTION: In a vision system configuration 100, a process calculates x-coordinates and y-coordinates in a gradient field in each image part in order to specify line features, projects the gradient field to a plurality of sub sections, detects a plurality of gradient extreme values, and creates a plurality of edge points having positions and gradients, repetitively selects two edge points, fits model lines to them and, when an edge point gradient coincides with a model, calculates a complete set of inlier points where a position and a gradient coincide with a model. A candidate line having a maximum inlier count is saved, and remaining set of outlier points is derived. Next, the process repetitively applies line fitting operation to the set and a subsequence outlier set, and detects a plurality of line results. The process may be comprehensive or be based on random sample consensus.SELECTED DRAWING: Figure 1

Description

RELATED APPLICATIONS This application claims the benefit of co-pending U.S. Provisional Patent Application Serial No. 62/249,918, entitled "System and Method for Detecting Lines in a Vision System," filed February 11, 2015, which is hereby incorporated by reference. The teachings are incorporated herein by reference.

FIELD OF THE INVENTION This invention relates to machine vision systems and, more particularly, to vision system tools for detecting line features in acquired images.

Machine vision systems (also referred to herein simply as "vision systems") are used for a variety of tasks in manufacturing, logistics, and industry. Such tasks include inspecting surfaces and parts, aligning objects in the assembly process, reading patterns and ID codes, and other operations where visual data is acquired and interpreted for use in subsequent processes. You can A vision system typically uses one or more cameras to acquire images of a scene containing objects or objects of interest. The bodies/objects may be stationary or in relative motion. Motion can also be controlled by information provided by a vision system, such as when a robot manipulates parts.

A common task in vision systems is to detect and characterize line features in images. Various tools are used to identify and analyze such line features. Typically, these tools rely on distinct contrast differences occurring in parts of the image. This contrast difference is analyzed using, for example, the caliper tool to determine if individual points in the image with contrast difference can be clustered into line-like features. If so, the lines are identified in the image. In particular, the tools for finding edge points and fitting lines to points work independently of each other. This increases processing overhead and reduces reliability. If the image contains multiple lines, such tools will be limited in their ability to accurately identify them. Additionally, tools designed to detect single lines in an image may be problematic to use when multiple lines of similar orientation and polarity are clustered together in the image.

SUMMARY OF THE INVENTION The present invention overcomes the shortcomings of the prior art by providing a system and method for detecting line features in an image that allows multiple lines to be efficiently and accurately identified and characterized. First, the process computes the x- and y-components of the gradient field at each image location, projects the gradient field onto multiple sub-regions, finds multiple gradient extrema within each sub-region, and identifies the associated position and gradients. Next, the process iteratively selects two edge points, fits a model line to those edge points, and if the slopes of those edge points match the model line, then the position and slope match the model line. Compute the complete set of inlier points The candidate line with the largest inline count is saved as the line result and the remaining set of outlier points is derived. The process then iteratively applies line fitting operations to this set and subsequent outline sets to detect multiple line results. The line fitting process may be exhaustive or based on random sample consensus (RANSAC).

According to an exemplary embodiment, a system is provided for detecting line features within an acquired image. A vision system processor receives image data of a scene containing line features. The edge point extractor generates an intensity gradient image from the image data and detects edge points based on the intensity gradient image. A line finder then fits the edge points to one or more lines based on the intensity gradients at the edge points. Illustratively, the line finder operates a RANSAC-based process to fit inlier edge points to new lines, which iteratively defines lines from outlier edge points relative to a pre-defined line. Including. The edge point extractor performs a projection of the area gradient field containing the line features of the intensity gradient image. Illustratively, the gradient field projections are oriented along directions set in response to expected orientations of one or more features or line features, and the gradient field projections have a granularity based on a Gaussian kernel. can be defined. Illustratively, the image data may include data from multiple images acquired from multiple cameras and transformed into a common coordinate space. The image data can also be smoothed using a smoothing (weighting) kernel, which may include a 1D Gaussian kernel or other weighting function. Edge points may be selected with thresholds defined by absolute contrast and normalized contrast based on the average intensity of the image data. Illustratively, the line finder may replace edge points representing portions of parallel or intersecting lines to correct for incorrect orientation and/or detect mixed polarities within the line based on gradient values at the edge points. constructed and arranged to identify lines having a polarity change of line features including . Also illustratively, the edge point extractor is arranged to detect a plurality of gradient maxima in each of the gradient projection sub-regions. Each of these gradient maxima can be identified as part of a plurality of edge points and can be described by a position vector and a gradient vector. Additionally, the line finder determines a match between at least one edge point of the plurality of extracted edge points and at least one candidate line of the plurality of detected lines by calculating a metric. can be placed. This metric may be based on the distance of at least one edge point from the candidate line and the angular difference between the gradient direction of the edge point and the normal direction of the candidate line.

The present invention will now be described with reference to the drawings.

FIG. 4 is a diagram of an exemplary vision system configuration and a vision system processor including edge detection tools/modules for acquiring an image of an object that includes multiple edge features, according to an example embodiment;

FIG. 1 illustrates an overall system and method for extracting edge points and detecting lines from an acquired image, according to an exemplary embodiment;

3 is a flowchart of an edge point extraction procedure according to the system and method of FIG. 2;

4 is a diagram of field projection onto an area of an image containing edge features, which is part of the edge point extraction procedure of FIG. 3; FIG.

Figure 4 illustrates applying a Gaussian kernel to an image to smooth the image for use in the edge point extraction procedure of Figure 3;

FIG. 4 is a field projection diagram including the application of a Gaussian kernel to smooth the projections used in the edge point extraction procedure of FIG. 3;

4 shows an overview of the edge point extraction procedure of FIG. 3, including application of a Gaussian kernel and calculation of absolute and normalized contrast thresholds for edge points; FIG.

FIG. 11 shows areas of contrast qualified for edge points with sufficient absolute and normalized contrast thresholds;

4 is a flowchart of a line detection procedure based on the edge points detected in FIG. 3 using an exemplary RANSAC, according to an exemplary embodiment;

FIG. 12 illustrates incorrect and correct alignment of edge points to dense parallel line features; FIG. 12 illustrates incorrect and correct alignment of edge points to dense parallel line features;

4A-4D illustrate correctly and incorrectly aligned edge points with respect to intersecting line features, as may be resolved by the line finder of the exemplary system and method, respectively; 4A-4D illustrate correctly and incorrectly aligned edge points with respect to intersecting line features, as may be resolved by the line finder of the exemplary system and method, respectively;

FIG. 10 illustrates light to dark polar, dark to light polar, light to dark or dark to light polar, or mixed polarities that may be resolved by the line finder of the exemplary systems and methods. FIG. 10 illustrates light to dark polar, dark to light polar, light to dark or dark to light polar, or mixed polarities that may be resolved by the line finder of the exemplary systems and methods. FIG. 10 illustrates light to dark polar, dark to light polar, light to dark or dark to light polar, or mixed polarities that may be resolved by the line finder of the exemplary systems and methods. FIG. 10 illustrates light to dark polar, dark to light polar, light to dark or dark to light polar, or mixed polarities that may be resolved by the line finder of the exemplary systems and methods.

FIG. 10 illustrates coverage score correction for detected lines considering a user-defined mask;

An exemplary vision system configuration 100 that can be used in accordance with exemplary embodiments is shown in FIG. System 100 includes at least one vision system camera 110, with one or more additional optional cameras 112 (shown in phantom). The exemplary cameras 110, 112 include an image sensor (or imager) S and associated electronics that acquire and transmit image frames to a vision system process (processor) 130, which is a stand-alone processor. and/or may be illustrated as computing device 140 . Cameras 110 (and 112) may include suitable lenses/optics 116 focused on the scene containing object 150 under inspection. Cameras 110 (and 112) may include internal and/or external illuminators (not shown) that operate according to the image acquisition process. Computing device 140 may be any acceptable processor-based system capable of storing and manipulating image data in accordance with an illustrative embodiment. For example, computing device 140 may include a PC (example shown), server, laptop, tablet, smart phone, or other similar device. Computing device 140 may include a bus-based image capture card that interconnects with suitable terminal equipment, such as a camera. In alternative embodiments, the vision processor can be partially or completely contained within the camera body itself and can be networked with other PCs, servers and/or camera-based processors to share and process image data. . Computing device 140 optionally includes a suitable display 142 capable of supporting a suitable graphical user interface (GUI) operable in accordance with vision system tools and processor 132 provided within vision system process (processor) 130 . can. Note that in various embodiments the display may be omitted and/or provided only for setup and service functions. The vision system tool may be part of any acceptable software and/or hardware package approved for use in inspecting objects, such as those commercially available from Cognex Corporation of Natick, Massachusetts, USA. good. The computing device may also include related user interface (UI) components including, for example, a keyboard 144 and mouse 146 as well as a touch screen within display 142 .

Cameras 110 (and 112) image some or all of objects 150 located within the scene. Each camera defines an optical axis OA around which a field of view is established based on optics 116, focal length, and the like. Object 150 includes a plurality of edges 152, 154 arranged in different directions. For example, an edge of an object may include an edge of a cover glass assembled within a smartphone body. Illustratively, the camera can image the entire object or a specific point (eg, the corner where the glass meets the body). The (common) coordinate space can be defined with reference to an object, a camera or some other reference point (eg a movable stage on which the object 150 rests). As shown, coordinate space is represented by axes 158 . These axes define illustratively orthogonal x-, y-, and z-axes, and a rotation Θz about the z-axis in the xy plane.

According to an exemplary embodiment, vision system processes 130 collectively interact with one or more applications/processes (running on computing device 140), including a set of vision system tools/processes 132. These tools cover a wide variety of conventional and specialized applications used to resolve image data - for example, a variety of calibration tools that can be used to transform acquired image data into a given (e.g. common) coordinate system. and may include affine transformation tools. A tool can also be included that converts the image grayscale intensity data to a binary image based on a predetermined threshold. Similarly, tools can be provided to analyze intensity gradients (contrast) between adjacent image pixels (and sub-pixels).

Vision system processes (processor) 130 includes a line detection process, tool, or module 13 that locates multiple lines within an acquired image according to an exemplary embodiment. Referring now to FIG. 2, an overview of line detection procedure 200 in accordance with an exemplary embodiment is illustrated. Procedure 200 consists of two main parts. An input image 210 is provided to the processor. As shown, the image includes a pair of intersecting edges 212 and 214 . These may represent the corner areas of object 150 described above. Edge point extractor 220 processes input image 210 to obtain a set of candidate edge points 230 including edge points 232 and 234 lying along edges 212 and 214, respectively. Edge points 232, 234 and their associated data (eg, intensity gradient information described below) are provided to recursive line finder 240, which performs a series of iterative processes on selected edge points. The goal of the iterative process is to attempt to fit other detected edge points to the candidate line features. Line detection process 240 results in detection of lines 252 and 254 as shown. These results may be provided to other downstream processes 260 that use the information - such as alignment processes, robotic manipulation, inspection, ID reading, part/surface inspection, and the like.

Refer to FIG. 3, which describes a procedure for extracting edge points, according to one embodiment. One or more images of a scene containing objects or surfaces having edge features to be detected are acquired (step 310). Images can be captured by a single camera or multiple cameras. In either case, at step 320 the image pixels may be (optionally) transformed into the new and/or common coordinate space by appropriate calibration parameters. This step may also include image smoothing, as described below. In embodiments in which multiple cameras image discrete areas of a scene--eg, focus on corner areas of larger objects--the common coordinate space can occupy empty areas between the camera fields of view. As will be explained below, a line extending between such fields of view (eg, an object edge connecting two detected corner areas) is determined by the system and method of an example embodiment. and method can be extrapolated. At step 330, the edge points needed to detect the lines are extracted from the image using gradient field projections by an edge point extractor in the appropriate coordinate space. Gradient values are first calculated for each pixel to generate an image for the x and y gradient components. The image is further processed by projecting the gradient field onto a number of caliper-like areas. Unlike conventional caliper tools that project intensity values, projecting gradient fields according to embodiments allows the orientation of gradients to be preserved, which facilitates subsequent line detection processes, as described below.

At step 340, referring also to the diagram of FIG. 4, the portion of the image (the caliper-like area) 400 containing the candidate edge features is subjected to gradient field projections (represented by a plurality of projections 410, 420, 430). received, searched over the (approximately) expected edge orientation in the search direction (arrow SD), and the projection is repeated over the area 400 in the orthogonal projection direction (arrow PD). For each projection (eg projection 420) the edge appears as a maximum in the gradient field 440 associated with the projection. Generally, a series of edge points within the projection associated with an edge exhibit an intensity gradient (vectors 552, 554) perpendicular to the direction of extension of the edge. As explained below, the user can define the projection direction based on the expected line orientation. Alternatively, this may be provided by a pre-determined mechanism or other mechanism - eg analysis of features in the image.

Two granularity parameters are involved in the gradient projection step described above. Prior to gradient field computation, the user can choose to smooth the image using an anisotropic Gaussian kernel. The first granularity determines the size of this Gaussian smoothing kernel. A Gaussian kernel of approximate size (eg, large 512, medium 514, small 516) is applied to smooth the image 210, as shown in diagram 500 of FIG. The first granularity parameter therefore determines the size of the isotropic Gaussian smoothing kernel prior to field computation.

This causes the process to perform a Gaussian weighted projection after the gradient field computation, rather than the uniform weighting in the conventional caliper tool. The second granularity parameter thus determines the size of the one-dimensional (1D) Gaussian kernel used during field projection, and area 600 is the Gaussian smoothing kernel 610, 620, 630 as shown in FIG. receive. During a typical operation, the user verifies (using the GUI) all the extracted edges superimposed on the image, and then the number of extracted edges along the line to be detected seems sufficient. while avoiding an excessive number of edges due to background noise in the image. In other words, this step allows the signal-to-noise ratio to be optimized for image features. This adjustment may also be performed automatically by the system using default values in various embodiments. Note that the Gaussian weighting function is one of a variety of strategies for weighting projections, including (for example) uniform weighting.

The overall flow of gradient field extraction and projection is illustrated in diagram 700 of FIG. Two granularity parameters, an isotropic Gaussian kernel 710 and a 1D Gaussian kernel 720 are shown in each half of the overview diagram 700 . As shown, each acquired image 210 undergoes smoothing and decimation 730 . The resulting image 740 then undergoes gradient field computation 750 as described above to produce two gradient images 752 and 754 . These gradient images are also denoted g _x and g _y , each representing two orthogonal axes in a common coordinate space. Note that in addition to the two gradient images, the intensity image 756 typically undergoes a smoothing, decimation and projection process 760 (using a Gaussian weighted projection 770 based on a 1D Gaussian kernel 720). This is because the processed intensity information is also used to calculate normalized contrast according to embodiments described below. The result is the projection profile of gradient images 772 (g _x ), 774 (g _y ) and intensity image 776 .

Referring also to step 350 of procedure 300 (FIG. 3), qualified edge points are then extracted by combining the 1D projection profiles of the x-gradient and y-gradient images. This is accomplished using a raw contrast calculation 780 and a normalized contrast calculation 790 based on the intensity image. More specifically, local peaks with both raw projected gradient magnitudes and normalized projected gradient magnitudes above a threshold, respectively, are detected in subsequent line detection according to the following exemplary formula: are considered candidate edge points for
(g _x ² +g _y ² ) ^1/2 >T _ABS
(g _x ² +g _y ² ) ^1/2 /I>T _NORM
where g _x and g _y are the values of the x and y gradient projections at the pixel location respectively, I is the intensity, T _AB is the absolute contrast threshold for the raw projected gradient magnitude, and T _NORM is the intensity normal is the normalized contrast threshold for the normalized projected gradient magnitude.

In particular, a point is considered a candidate edge point only if its absolute contrast and normalized contrast both exceed their respective thresholds. This is illustrated by the upper right quadrant 810 in the exemplary graph 800 of contrast threshold T _ABS against normalized contrast threshold T _NORM . Using dual thresholds (absolute and normalized) is totally different from existing strategies that typically use absolute contrast thresholds. The advantage of dual contrast thresholding is evident, for example, when an image contains both dark and light intensity areas, both of which contain edges of interest. To detect edges in dark areas of an image, it is desirable to set a low contrast threshold. However, setting such a low contrast may result in detecting false edges in bright areas of the image. Conversely, it is desirable to set a high contrast threshold to avoid detecting false edges in bright areas of the image. However, with high contrast settings, the system may not be able to detect edges properly in dark areas of the image. By using a second normalized contrast threshold in addition to the traditional absolute contrast threshold, the system is able to detect edges properly in both dark and bright areas, while avoiding erroneous edges in bright areas of the image. Edge detection can be avoided. Therefore, the use of dual contrast thresholding maximizes the speed and robustness of subsequent line detection stages of the overall process, while avoiding false edges while allowing relevant edge detection. Helpful.

Still referring to procedural step 350 (FIG. 3), once all edge points have been extracted, they are reconstructed and stored in a convenient data structure for subsequent line finder operations. For example, note the following tuple:
p = (x, y, gx, gy, gm, go, I, gm/I, m, n)
where (x, y) is the location of the edge point, (g _x , g _y ) are the values of the x and y gradient projections respectively, and (g _m , g _o ) are calculated from (g _x , g _y ) is the intensity at the edge point, g _m /I is the intensity-normalized contrast obtained by dividing the gradient magnitude g _m by the intensity I, m is the image index, and n is the projected area index. Edge point locations, such as in the standard caliper tool, can be interpolated to improve accuracy.

Note that the edge point extraction process generally operates by projecting fields in a single direction that substantially matches the expected line angle. The tool is therefore most sensitive to edges at this angle, and its sensitivity gradually decreases for edges at other angles, the rate of decrease indirectly determining the length of the field projection. depends on the setting of As a result, provided that the angular range is specified by the user, the process is limited to finding lines whose angles are "close to" the expected line angle. Although the process is adapted to detect non-orthogonal lines, in various embodiments any angle over 360 degrees can be detected by performing projections in multiple directions, including orthogonal directions (omnidirectional line detection). is assumed to be generalized to detect lines of

Referring to step 360 of procedure 300 (FIG. 3), edge point candidates that exceed a threshold are provided to a line finder in accordance with an illustrative embodiment. By way of example, the line finder operates recursively and employs (for example) random sample consensus (RANSAC) based techniques. See also line detection procedure 900 in FIG. At step 910, the user inputs the maximum number of lines expected in an image along with the expected angle, angle error, distance error and (illustratively) a minimum coverage score (generally defined below) via (for example) a GUI specified by These parameters are used by the linefinder to manipulate the following processes. Lines for each subregion of the image are found by running the RANSAC line finder recursively, with edge point outliers from one stage becoming input points for the next stage. Thus, at step 920, procedure 900 selects a pair of edge points that are part of the group of edge points identified as extrema in the edge detection process. Procedure 900 attempts to fit a model line to selected edge points based on matching gradient values that are consistent (within a selected error range) with the model line. One or more line candidates from step 922 are returned in step 924 . Each line detection step returns a candidate line, its inliers and outliers. The returned lines are computed for inlier edge points that have locations and slopes that match the line candidates (step 926). At step 928, the candidate line with the largest inline count is identified. The line detection steps described above (steps 920-928) terminate when the maximum number of allowed RANSAC iterations is reached (decision step 930). The maximum number of iterations within each line detection stage is automatically calculated using the internally calculated worst case proportion of outliers and the user-specified assurance level. Each line detection stage returns from all its iterations the line with the maximum number of camptured edge points, subject to user-specified fitting error, geometric constraints and polarity. Each edge point can be assigned to only one line inline list, and each line is allowed to contain at most one edge point from each projection area. The gradient orientation of an edge point, along with its position, is used to determine whether it should be included in the candidate line's inlier list. Specifically, the edge point should have a gradient orientation that matches the angle of the candidate line.

Decision step 930 determines that more iterations are allowed to return outliers from the best inlier candidates (step 940) for use by the RANSAC process (step 920) in detecting line candidates.

In each RANSAC iteration, two edge points belonging to different projection areas are randomly selected and a line is fitted to those two points. The resulting candidate line is further considered only if its angle matches the slope angles of both edges of the point pair and if the line angle matches a user-specified uncertainty range. In general, edge point gradient directions are usually orthogonal, but are allowed to differ by a user-defined angular error. If the candidate line passes these initial tests, then the number of inlier edge points is evaluated, otherwise a new RANSAC iteration is started. An edge point is considered an inlier of a candidate line only if its gradient direction and position match the line based on the gradient angle and distance error specified by the user.

Once the maximum number of RANSAC iterations has been reached (decision step 930), the best line candidate inliers found are subjected to improved line fitting using (for example) least squares regression or other acceptable approximation methods. Then the set of inlier edge points is re-evaluated and these steps are repeated up to N times (eg, 3 or more times) until the number of inliers no longer increases or decreases (step 960). This is the line output as the detected line in step 970 .

Decision step 980 determines if more lines should be found (by searching (for example) for more subregions or other criteria), and if so the process loops back to step 920 for a new set of (step 982). Once points are exhausted or the maximum number of iteration counts is reached, procedure 900 returns the set of lines detected (in multiple images) at step 990 .

The multi-line finder is adapted to perform a final adjustment of existing results when two lines intersect each other within the inspection area. As generally shown in FIGS. 10 and 11, for closely spaced parallel lines 1010 and 1020, erroneous line results (i.e., FIG. 10) are sometimes obtained due to the statistical accuracy of the RANSAC procedure. can be However, when such an error occurs, an exchange of inlier point groups (arrow 1120 in group 1110 in FIG. 11) can sometimes locate the correct line with increased coverage scores and decreased fitting residuals. Point swapping is very effective when the image contains closely spaced parallel lines, as shown. Conversely, if the image contains lines 1210 and 1220 that actually intersect each other as illustrated in FIGS. 12 and 13, the coverage score is reduced after redemption (arrow 1230 ), the first result obtained before the exchange is preserved by the process of successfully detecting the crossing line.

Note that the RANSAC procedure is one of a variety of techniques by which a line finder can fit points to lines. In alternative embodiments, candidate points can be selected according to a displacement set between them, or the image can be processed using (for example) an exhaustive search technique. Accordingly, references to the RANSAC technique as used herein should be taken to include a wide variety of similar point-fitting techniques.

Additional features of this system and method may be provided. These include support for mixed polarities, automatic calculation of projection area width, support for multi-view line detection, and allowing distortion-free input images to remove optical distortions. These functions are described below.

Still referring to the examples of FIGS. 14-16, the line detection systems and methods of the illustrative embodiments generally provide standard bright-to-dark, dark-to-light and (respectively) contrasts for contrast between detected edges. Supports alternate polarity settings. In addition, the system and method can also support mixed polarity settings (FIG. 17) where both light-to-dark and dark-to-light features appear on the same line. The line detection results for all four settings are shown in the following figures. In an exemplary embodiment, the system and method may include a mixed polarity setting that allows detection of a single line containing edge points of opposite polarity. This differs from the conventional "alternative" polarity setting, where all edge points of a line are of either polarity, but are limited to one polarity. Mixed polarity settings may be advantageously used for (for example) light-dark graticules in calibration plates, among other applications.

The user can select improved shift invariant line detection. In such cases, the edge point extractor uses substantially overlapping projected areas to improve the stability of the results. If these areas do not overlap, the pixel of interest may fall outside the projection area when the image shifts, resulting in poor shift invariance in line detection results. Overlapping projection areas ensure that the pixels of interest are continuously covered by the projection areas. If overlapping projection areas are used, incremental computations can be performed with possible low-level optimizations.

A user may provide a mask that excludes certain portions of the acquired image and/or imaged surface from the analysis of line features. This may be desirable if the surface contains known line features of no interest (e.g. barcodes, text, and other structures that are not closely related to the task of trying to detect lines that are parsed by other mechanisms). be. Thus, the edge point extractor can support image masking where "irrelevant" areas in the image can be masked out and "relevant" areas can be masked in. Once such masking occurs, the coverage score of the exemplary detected line is reweighted according to the number of edge points falling within the mask.

Referring to exemplary image area 1800 in FIG. 18, it shows coverage scores in the presence of image masks and the effect of image masking on such coverage scores. The edge point extractor supports image masking that allows masking out "irrelevant" areas in the image. As shown, the detected line 1810 is characterized by relational edge points (based on the "relationship" mask area 1820). Such related edge points consist of related edge point inliers 1830 for line 1810 and irrelevant edge point outliers 1840 for line 1810 . An irrelevant edge point 1850 on line 1810 is between the relevant areas 1820 of the mask as shown in this example and is not included in the coverage score calculation even though it is on the line as an inlier. Potential locations 1860 for edge points along line 1810 have also been determined as shown. These potential locations are located between known points at predictable intervals based on the spacing of the detected points. Illustratively, the coverage score of the detected line is reweighted according to the number of edge points falling within the mask. This modifies the coverage score as follows:
Coverage score = number of related edge point inliers for line/(number of related edge point inliers for line + number of related edge point outliers for line + number of potential locations of related edge points).

After performing the line detection process according to the systems and methods described herein, the detected lines can be classified in a variety of ways based on user-specified classification criteria (eg, via a GUI). The user can select from internal classifiers such as in-line coverage score, intensity or contrast. The user can also select from internal classifiers, such as signed distances or relative angles. When using an external classifier, the user can specify a reference line segment for calculating the external classifier for the detected line.

As described generally above, the system and method can include a multi-field-of-view (MFOV) overload, allowing vectors of images from different fields of view to be put into the process. All images should be in the common client coordinate space based on calibration. As noted above, this functionality can be extremely useful in application scenarios where multiple cameras are used to capture sub-regions of a single part. Since edge points carry gradient information, line features projected between gaps in the field of view can still be resolved (when the gradients in both FOVs are comparable to the line orientation and alignment given in each FOV). .

In particular, the system and method do not require distortion removal (i.e., the image is not distorted) to remove non-linear distortion and ensure that the distortion is not significant (allowing for undistorted images). do). If the image is not distorted, the system and method can still detect candidate edge points and map point locations and gradient vectors through non-linear transformations.

It will be apparent that the system and method and line finders provided in accordance with various alternative embodiments/improvements are effective and robust tools for determining multiple line features under a variety of conditions. Generally, when used to detect line features, the system and method have no particular limit on the maximum number of lines to be detected in an image. Only storage and computation time will impose a practical limit on the number of lines that can be detected.

Exemplary embodiments of the invention are described above in detail. Various modifications and additions can be made without departing from the spirit and scope of the invention. Features of each of the various embodiments described above may be combined with features of another described embodiment so long as it is suitable to provide multiple feature combinations in the associated new embodiment. Moreover, although numerous separate embodiments of the apparatus and method of this invention have been described above, those described herein are merely illustrative of the application of the principles of this invention. For example, the terms "process" and/or "processor" as used herein are used broadly to refer to various electronic hardware- and/or software-based functions and components (alternatively functional "modules" or "elements"). ) should be interpreted as including Additionally, any illustrated process or processor may be combined with other processes and/or processors or divided into various sub-processes or sub-processors. Such sub-processes and/or sub-processors can be variously combined according to the embodiments described herein. Likewise, it is expressly understood that any function, process and/or processor herein can be implemented using electronic hardware comprising a non-transitory computer-readable medium of program instructions, software, or a combination of hardware and software. It is assumed. Further, terms representing various directions and/or orientations as used herein, such as "vertical", "horizontal", "top", "bottom", "bottom", "top", "side" , ``front'', ``back'', ``left'', ``right'' and the like are used only as relative expressions and refer to a fixed coordinate system such as the direction of gravity. It does not represent absolute orientation. Accordingly, this description should be taken as illustrative only and is not meant to otherwise limit the scope of the invention. Additionally, when the words "substantially" or "approximately" are used in reference to a given measurement, value or characteristic, it refers to the amount within the normal operating range to achieve the desired result. but includes some variation (eg, 1-5 percent) due to inherent inaccuracies and errors within the tolerance of the system. Accordingly, this description is by way of example and is not meant to otherwise limit the scope of the invention.

Claims

Claims (18)

A system for detecting line features in an acquired image, comprising:
The system comprises a vision system processor that receives an acquired image of a scene and a line finder,
The vision system processor above
(a) subjecting the acquired images to a gradient field computation process to generate two gradient images and an intensity image;
(b) applying weighted projections to the two gradient images and an intensity image to obtain a one-dimensional (1D) projection profile of the two gradient field and projection images;
(c) extracting candidate edge points by combining the 1D projection profiles of the two gradient images;
wherein the line finder generates a plurality of lines that coincide with the extracted edge points;
the above system. 2. The system of claim 1, wherein the vision system processor generates raw contrast magnitudes and normalized contrast magnitudes based on 1D projection profiles of the two gradient images. 3. The system of claim 2, wherein said normalized contrast magnitude is based on said two gradient fields and an intensity image. 4. The system of claim 3, wherein the candidate edge points each have a raw contrast magnitude and a normalized contrast magnitude that exceed a threshold. 2. The system of claim 1, wherein prior to said (a), said vision system processor smoothes and decimates said acquired image. 2. The system of claim 1, wherein said weighted projections are uniformly weighted projections. 2. The system of claim 1, wherein said weighted projections are Gaussian weighted projections. 2. The system of claim 1, wherein said weighted projection is based on a one-dimensional Gaussian kernel. 2. The system of claim 1, wherein said two gradient images represent two orthogonal axes in a common coordinate space. A method for detecting line features in an acquired image, comprising:
Receive an image of the acquired scene,
subjecting the acquired images to a gradient field computation process to generate two gradient images and an intensity image;
applying a weighted projection to the two gradient images and an intensity image to obtain a one-dimensional (1D) projection profile of the two gradient field and projection images;
extracting candidate edge points by combining the 1D projection profiles of the two gradient images, and generating a plurality of lines that match the extracted edge points;
above method. 11. The method of claim 10, further comprising generating a raw contrast magnitude and a normalized contrast magnitude based on 1D projection profiles of the two gradient images. 12. The method of claim 11, wherein said normalized contrast magnitude is based on said two gradient fields and an intensity image. 13. The method of claim 12, wherein the candidate edge points each have a raw contrast magnitude and a normalized contrast magnitude that exceed a threshold. 11. The method of claim 10, wherein (a) is preceded by smoothing and decimation of the acquired image. 11. The method of claim 10, wherein said weighted projections are uniformly weighted projections. 11. The method of claim 10, wherein said weighted projections are Gaussian weighted projections. 11. The method of claim 10, wherein said weighted projection is based on a one-dimensional Gaussian kernel. 11. The method of claim 10, wherein said two gradient images represent two orthogonal axes in a common coordinate space.

Images