Feature Detection and Extraction

Image registration, interest point detection, feature descriptor extraction, point feature matching, and image retrieval

Local features and their descriptors are the building blocks of many computer vision algorithms. Their applications include image registration, object detection and classification, tracking, motion estimation, and content-based image retrieval (CBIR). These algorithms use local features to better handle scale changes, rotation, and occlusion. Computer Vision Toolbox™ algorithms include the FAST, Harris, and Shi & Tomasi corner detectors, and the SIFT, SURF, KAZE, and MSER blob detectors. The toolbox includes the SIFT, SURF, FREAK, BRISK, LBP, ORB, and HOG descriptors. You can mix and match the detectors and the descriptors depending on the requirements of your application.

Apps

Registration EstimatorRegister 2-D grayscale images

Functions

expand all

detectBRISKFeaturesDetect BRISK features
detectFASTFeaturesDetect corners using FAST algorithm
detectHarrisFeaturesDetect corners using Harris–Stephens algorithm
detectKAZEFeaturesDetect KAZE features
detectMinEigenFeaturesDetect corners using minimum eigenvalue algorithm
detectMSERFeaturesDetect MSER features
detectORBFeaturesDetect ORB keypoints
detectSIFTFeaturesDetect scale invariant feature transform (SIFT) features (Since R2021b)
detectSURFFeaturesDetect SURF features
extractFeaturesExtract interest point descriptors
extractLBPFeaturesExtract local binary pattern (LBP) features
extractHOGFeaturesExtract histogram of oriented gradients (HOG) features
matchFeaturesFind matching features
matchFeaturesInRadiusFind matching features within specified radius (Since R2021a)
estgeotform2dEstimate 2-D geometric transformation from matching point pairs (Since R2022b)
estgeotform3dEstimate 3-D geometric transformation from matching point pairs (Since R2022b)
imwarpApply geometric transformation to image
imblendBlend two images (Since R2024b)
vision.BlockMatcherEstimate motion between images or video frames
vision.TemplateMatcherLocate template in image
insertMarkerInsert markers in image or video
insertShapeInsert shapes in image or video
showMatchedFeaturesDisplay corresponding feature points
showShapeDisplay shapes on image, video, or point cloud (Since R2020b)
insertObjectAnnotationAnnotate truecolor or grayscale image or video
insertObjectKeypointsInsert object keypoints in image (Since R2023b)
insertTextInsert text in image or video
imshowDisplay image
imshowpairCompare differences between images
vision.ChromaResamplerDownsample or upsample chrominance components of images
binaryFeaturesObject for storing binary feature vectors
BRISKPointsObject for storing BRISK interest points
cornerPointsObject for storing corner points
KAZEPointsObject for storing KAZE interest points
MSERRegionsObject for storing MSER regions
ORBPointsObject for storing ORB keypoints
SIFTPointsObject for storing SIFT interest points (Since R2021b)
SURFPointsObject for storing SURF interest points
rigidtform2d2-D rigid geometric transformation (Since R2022b)
simtform2d2-D similarity geometric transformation (Since R2022b)
affinetform2d2-D affine geometric transformation (Since R2022b)
projtform2d2-D projective geometric transformation (Since R2022b)
rigidtform3d3-D rigid geometric transformation (Since R2022b)
simtform3d3-D similarity geometric transformation (Since R2022b)

Create Recognition Database

bagOfFeaturesBag of visual words object
invertedImageIndexSearch index that maps visual words to images

Retrieve Images

retrieveImagesSearch image set for similar image
imageDatastoreDatastore for image data
evaluateImageRetrievalEvaluate image search results

Topics

Featured Examples

Image Retrieval Using Customized Bag of Features

Image Retrieval Using Customized Bag of Features

Create a Content Based Image Retrieval (CBIR) system using a customized bag-of-features workflow.

Pattern Matching

Pattern Matching

Use the 2-D normalized cross-correlation for pattern matching and target tracking. The example uses predefined or user specified target and number of similar targets to be tracked. The normalized cross correlation plot shows that when the value exceeds the set threshold, the target is identified.

Find Object in Cluttered Scene Using Image Point Features

Find Object in Cluttered Scene Using Image Point Features

Detect a particular object in a cluttered scene, given a reference image of the object.

Digit Classification Using HOG Features

Digit Classification Using HOG Features

Classify digits using HOG features and a multiclass SVM classifier.

Automatically Find Image Rotation and Scale

Automatically Find Image Rotation and Scale

Demonstrates how to automatically determine the geometric transformation between two images. Specifically, when one image is distorted in relation to another due to rotation and scaling, the functions detectSURFFeatures and estimateGeometricTransform2D can be employed to identify the rotation angle and scale factor. Subsequently, these parameters can be used to transform the distorted image back to its original appearance.

Create Panorama

Create Panorama

Automatically stitch multiple images into panorama. The procedure for image stitching is an extension of feature based image registration. Instead of registering a single pair of images, multiple image pairs are successively registered relative to each other to form a panorama.

Stabilize Video Using Image Point Features

Stabilize Video Using Image Point Features

Stabilize a video that was captured from a jittery platform. One way to stabilize a video is to track a salient feature in the image and use this as an anchor point to cancel out all perturbations relative to it. This procedure, however, must be bootstrapped with knowledge of where such a salient feature lies in the first video frame. In this example, we explore a method of video stabilization that works without any such a priori knowledge. It instead automatically searches for the "background plane" in a video sequence, and uses the observed distortion to correct for camera motion.

Object Counting

Object Counting

Use morphological operations to count objects in a video stream.

Cell Counting

Cell Counting

Use a combination of basic morphological operators and blob analysis to extract information from a video stream. In this case, the example counts the number of E. Coli bacteria in each video frame. Note that the cells are of varying brightness, which makes the task of segmentation more challenging.