--- title: "ImageLines" language: "en" type: "Symbol" summary: "ImageLines[image] finds line segments in image and returns the coordinates of their endpoints. ImageLines[image, t] uses the threshold t for selecting image lines. ImageLines[image, t, d] uses the parameter d to control the distinctness of the detected lines. ImageLines[video, ...] finds lines in frames of video." keywords: - line detection - hough - edge detection - ransac - line segments - houghlines - houghpeaks canonical_url: "https://reference.wolfram.com/language/ref/ImageLines.html" source: "Wolfram Language Documentation" related_guides: - title: "Feature Detection" link: "https://reference.wolfram.com/language/guide/FeatureDetection.en.md" - title: "Image Processing & Analysis" link: "https://reference.wolfram.com/language/guide/ImageProcessing.en.md" - title: "Computer Vision" link: "https://reference.wolfram.com/language/guide/ComputerVision.en.md" - title: "Video Analysis" link: "https://reference.wolfram.com/language/guide/VideoAnalysis.en.md" - title: "Video Computation: Update History" link: "https://reference.wolfram.com/language/guide/VideoComputation-UpdateHistory.en.md" - title: "Image Computation: Update History" link: "https://reference.wolfram.com/language/guide/ImageComputation-UpdateHistory.en.md" - title: "Computational Photography" link: "https://reference.wolfram.com/language/guide/ComputationalPhotography.en.md" related_functions: - title: "Radon" link: "https://reference.wolfram.com/language/ref/Radon.en.md" - title: "FindImageShapes" link: "https://reference.wolfram.com/language/ref/FindImageShapes.en.md" - title: "EdgeDetect" link: "https://reference.wolfram.com/language/ref/EdgeDetect.en.md" - title: "GradientFilter" link: "https://reference.wolfram.com/language/ref/GradientFilter.en.md" - title: "ImageKeypoints" link: "https://reference.wolfram.com/language/ref/ImageKeypoints.en.md" - title: "ImageCorners" link: "https://reference.wolfram.com/language/ref/ImageCorners.en.md" - title: "HighlightImage" link: "https://reference.wolfram.com/language/ref/HighlightImage.en.md" - title: "HighlightVideo" link: "https://reference.wolfram.com/language/ref/HighlightVideo.en.md" --- # ImageLines ImageLines[image] finds line segments in image and returns the coordinates of their endpoints. ImageLines[image, t] uses the threshold t for selecting image lines. ImageLines[image, t, d] uses the parameter d to control the distinctness of the detected lines. ImageLines[video, …] finds lines in frames of video. ## Details and Options * ``ImageLines`` returns a list of line segments in the form ``Line[{p1, p2}]``, where each ``pi = {xi, yi}`` is expressed in the standard image coordinate system. * ``ImageLines[image, t]`` finds lines in the ``image`` whose normalized strength is larger than the specified threshold ``t``. * ``ImageLines`` sorts the result based on the normalized strength. * In ``ImageLines[image, t, d]``, the parameter ``d`` controls how close lines are suppressed. If the value is set to zero, all detected lines are returned. With ``d`` set to ``1``, only the strongest line may be returned. * The following options can be given: | | | | | ------------ | ------- | ------------------------------------ | | MaxFeatures | All | maximum number of features to return | | Method | "Hough" | method to detect lines | * With a setting ``MaxFeatures -> n``, at most ``n`` lines with largest normalized strength are returned. * Possible line detection methods are: | | | | -------- | ---------------------------------------- | | "Hough" | lines based on Hough transform (default) | | "RANSAC" | lines using the RANSAC algorithm | * With ``Method -> "Hough"``, lines are detected by iteratively selecting the strongest peaks in the Hough transform. Using the distinctness parameter, peaks that are within a rectangular range from the already selected peaks are excluded from the set of line candidates. * With ``Method -> "RANSAC"``, lines are detected using random sampling. For each sampling, pixels that are within a distance specified by the distinctness parameter ``d`` are used for computing the strength of the line. The pixels on the selected line are not used in the following iterations. * By default, ``ImageLines`` returns lines that span from border to border. With a setting ``Method -> {"Segmented" -> True}``, detected lines may be divided into smaller line segments. --- ## Examples (14) ### Basic Examples (1) Detect and visualize straight lines in an image: ```wl In[1]:= i = [image]; lines = ImageLines[i, 1 / 3] Out[1]= {Line[{{0., 172.014}, {360., 76.6467}}], Line[{{0., 106.159}, {360., 168.092}}], Line[{{0., 118.373}, {360., 180.306}}], Line[{{0., 123.596}, {320.562, 0.}}], Line[{{0., 90.4493}, {360., 157.446}}], Line[{{0., 49.5606}, {360., 126.765}}], Line[{{0., 72.8608}, {360., 142.399}}], Line[{{0., 130.942}, {360., 187.834}}], Line[{{0., 198.926}, {360., 126.839}}], Line[{{0., 24.3074}, {360., 106.659}}]} ``` Highlight detected lines: ```wl In[2]:= HighlightImage[i, lines] Out[2]= [image] ``` ### Scope (2) Specify the distinctness parameter: ```wl In[1]:= ImageLines[[image], .2, .2] Out[1]= {Line[{{0., 18.3048}, {201., 18.3048}}], Line[{{0., 118.107}, {118.82, 0.}}], Line[{{101.363, 120.}, {100.641, 0.}}]} ``` --- Detect the line going through the foreground pixels: ```wl In[1]:= ImageLines[[image], .05] Out[1]= {Line[{{0.492551, 40.}, {39.5074, 0.}}]} ``` ### Options (3) #### MaxFeatures (1) Specify the maximum number of lines to be to be detected: ```wl In[1]:= img = [image]; lines = ImageLines[EdgeDetect[img], MaxFeatures -> 5] Out[1]= {Line[{{71.4974, 225.}, {262.715, 0.}}], Line[{{232.978, 225.}, {13.8344, 0.}}], Line[{{51.8704, 225.}, {295.412, 0.}}], Line[{{0., 134.307}, {300., 138.827}}], Line[{{219.337, 225.}, {33.8854, 0.}}]} In[2]:= HighlightImage[img, {Orange, lines}] Out[2]= [image] ``` #### Method (2) By default, lines are detected using ``Method -> "Hough"`` : ```wl In[1]:= img = [image]; lines = ImageLines[img, .23, Method -> "Hough"]; HighlightImage[img, lines] Out[1]= [image] ``` Use a random sampling method: ```wl In[2]:= lines = ImageLines[img, .23, Method -> "RANSAC"]; HighlightImage[img, lines] Out[2]= [image] ``` --- Detect segmented lines in a grayscale image: ```wl In[1]:= img = [image]; lines = ImageLines[img, .2, Method -> {"Hough", "Segmented" -> True}]; HighlightImage[img, lines, ImageSize -> 240] Out[1]= [image] ``` Segmented lines using the random sampling method: ```wl In[2]:= lines = ImageLines[img, .23, Method -> {"RANSAC", "Segmented" -> True}];HighlightImage[img, lines, ImageSize -> 240] Out[2]= [image] ``` ### Applications (6) Detect and visualize straight trajectories in a bubble chamber image: ```wl In[1]:= img = [image]; lines = ImageLines[img]; HighlightImage[img, lines] Out[1]= [image] ``` --- Detect segments on a color image: ```wl In[1]:= img = [image]; lines = ImageLines[img, .53, .03, Method -> {"Segmented" -> True}]; HighlightImage[img, {Black, lines}, ImageSize -> 320] Out[1]= [image] ``` --- Visualize vanishing points: ```wl In[1]:= img = [image]; lines = ImageLines[EdgeDetect[img], .28, .06]; HighlightImage[img, lines] Out[1]= [image] ``` --- Detect segments on a gradient magnitude map: ```wl In[1]:= img = [image]; lines = ImageLines[ImageAdjust@GradientFilter[img, 3, Method -> {"NonMaxSuppression" -> True}], .09, Method -> {"Segmented" -> True}]; HighlightImage[img, {Yellow, lines}] Out[1]= [image] ``` --- Find wide lines using edge detection: ```wl In[1]:= img = [image]; lines = ImageLines[EdgeDetect[img, 13], .18]; HighlightImage[img, lines] Out[1]= [image] ``` --- Straighten the tower of Pisa: ```wl In[1]:= image = [image]; ``` Compute the gradient of the image to highlight edges: ```wl In[2]:= (g = GradientFilter[image, 2])//ImageAdjust Out[2]= [image] ``` Find the most significant straight lines in the gradient image: ```wl In[3]:= HighlightImage[image, lines = ImageLines[g, MaxFeatures -> 10]] Out[3]= [image] ``` FInd the angles corresponding to each line: ```wl In[4]:= angles = Mod[ArcTan@@@Flatten[Differences@@@lines, 1], π] Out[4]= {9.473903143468003`*^-17, 0.0866103, 0.0157473, 3.07073, 1.55899, 0.070863, 3.02349, 3.09435, 1.50387, 1.51174} ``` Compute the average angle of the vertical lines: ```wl In[5]:= θ = Mean@Select[angles, Abs[# - π / 2] < π / 4&] Out[5]= 1.52487 ``` Rotate the image by $\frac{\pi }{2}-\theta$ to make almost-vertical lines vertical: ```wl In[6]:= ImageRotate[image, π / 2 - θ] Out[6]= [image] ``` ### Possible Issues (2) Use the distinctness parameter to prevent detecting duplicated lines: ```wl In[1]:= Length@ImageLines[[image], .1, #]& /@ {0, .1} Out[1]= {72, 3} ``` --- Thin lines in binary images might not be correctly detected: ```wl In[1]:= i = [image]; In[2]:= lines = ImageLines[i, MaxFeatures -> 3, Method -> {"Segmented" -> True}]; HighlightImage[i, lines] Out[2]= [image] ``` Blurring the image typically improves line detection: ```wl In[3]:= lines = ImageLines[Blur[i], MaxFeatures -> 3, Method -> {"Segmented" -> True}]; HighlightImage[i, lines] Out[3]= [image] ``` Rescaling the image to a larger size might also help in some cases: ```wl In[4]:= i2 = ImageResize[i, Scaled[2]]; In[5]:= lines = ImageLines[i2, MaxFeatures -> 3, Method -> {"Segmented" -> True}]; HighlightImage[i2, lines] Out[5]= [image] ``` ## See Also * [`Radon`](https://reference.wolfram.com/language/ref/Radon.en.md) * [`FindImageShapes`](https://reference.wolfram.com/language/ref/FindImageShapes.en.md) * [`EdgeDetect`](https://reference.wolfram.com/language/ref/EdgeDetect.en.md) * [`GradientFilter`](https://reference.wolfram.com/language/ref/GradientFilter.en.md) * [`ImageKeypoints`](https://reference.wolfram.com/language/ref/ImageKeypoints.en.md) * [`ImageCorners`](https://reference.wolfram.com/language/ref/ImageCorners.en.md) * [`HighlightImage`](https://reference.wolfram.com/language/ref/HighlightImage.en.md) * [`HighlightVideo`](https://reference.wolfram.com/language/ref/HighlightVideo.en.md) ## Related Guides * [Feature Detection](https://reference.wolfram.com/language/guide/FeatureDetection.en.md) * [Image Processing & Analysis](https://reference.wolfram.com/language/guide/ImageProcessing.en.md) * [Computer Vision](https://reference.wolfram.com/language/guide/ComputerVision.en.md) * [Video Analysis](https://reference.wolfram.com/language/guide/VideoAnalysis.en.md) * [Video Computation: Update History](https://reference.wolfram.com/language/guide/VideoComputation-UpdateHistory.en.md) * [Image Computation: Update History](https://reference.wolfram.com/language/guide/ImageComputation-UpdateHistory.en.md) * [Computational Photography](https://reference.wolfram.com/language/guide/ComputationalPhotography.en.md) ## History * [Introduced in 2010 (8.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn80.en.md) \| [Updated in 2012 (9.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn90.en.md) ▪ [2014 (10.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn100.en.md) ▪ [2018 (11.3)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn113.en.md) ▪ [2025 (14.2)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn142.en.md)