Questions tagged [local-features]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
74
votes
4answers
57k views

What are some free alternatives to SIFT/ SURF that can be used in commercial applications?

As far as I understand, both SURF and SIFT are patent protected. Are there any alternative methods that can be used in a commercial application freely? For more info on the patent check out: http://...
17
votes
5answers
19k views

Scale and Rotation invariant feature descriptors

Can you list some scale and rotational invariant feature descriptors for use in feature detection. The application is for the detection of cars and humans in video captured by a UAV, using a multi-...
15
votes
1answer
5k views

Binary robust independent elementary features (“BRIEF”) understanding

How are the descriptors in the BRIEF algorithm matched to each other? How is the template image found in the another image? How do we compare the descriptors? I have read that article but didn't ...
15
votes
1answer
4k views

Detection of circles (ellipses) in 2D point cloud

Given a set of points (2D) i.e., point cloud (PC), the question is about a robust, accurate and ...
15
votes
2answers
1k views

Image Registration by Segmentation

Image registration algorithms are usually based on point features such as SIFT (Scale-Invariant Feature Transform). I saw some references to line features, but I was wondering if it would be possible ...
14
votes
1answer
18k views

What is the difference between feature detectors and feature descriptors?

What is the difference between feature detectors and feature descriptors? Which among these are detectors and which are descriptors: Harris, SURF, Min Eigen, FAST, SIFT, BRISK
10
votes
2answers
19k views

How Hessian feature detector works?

I know about Harris corner detector, and I understand the basic idea of its second moment matrix, $$M = \left[ \begin{array}{cc} I_x^2 & I_xI_y \\ I_xI_y & I_y^2 \end{array} \right]$$, edges ...
9
votes
4answers
3k views

What Kind of Features Can I Extract from a Signal

I want to monitor (automatic-)gearbox failures on some vehicles. For each vehicle I have a captured signal representing the selected gear at each one millisecond. An example of two signals are shown ...
9
votes
1answer
2k views

Good features / algorithms for recognizing car-models in images

I have a question concerning object recognition, especially recognizing car-models! I am at the beginning of a work about identifying the same car-model in different images. At the moment I think one ...
9
votes
2answers
1k views

Image Processing Prior to Feature Detection

I have implemented a feature detector based on Harris corners. It works fine most times, but there are cases where it performs poorly. I need to make it work on many different images without ...
8
votes
4answers
1k views

Preprocessing to improve Harris corner tracking between video frames?

In our unmanned aerial vehicle grayscale video image stabilization application, we're having difficulty finding the "good" Harris corners in frame N+1 selected from frame N. The source of the ...
8
votes
3answers
3k views

Purpose of image feature detection and matching

I'm a new guy in image processing and computer vision, so this question might be stupid to you. I just learned some feature detection and description algorithms, such as Harris, Hessian, SIFT, SURF, ...
8
votes
1answer
14k views

Feature extraction/reduction using DWT

For a given time series which is n timestamps in length, we can take Discrete Wavelet Transform (using 'Haar' wavelets), then we get (for an example, in Python) - ...
8
votes
1answer
1k views

Suggested Preprocessing methods for OCR on Circular Images

Hello this is my sample image I am going to do real time character detection on images like that. I've tried SURF, SIFT, MSER and template matching on original image without any preprocessing. I can ...
5
votes
1answer
4k views

difference between Gabor and log-Gabor function

I am reading a paper using log-gabor filters for feature detection. I was thinking about the difference between Gabor filters and log-gabor filters. Can anyone tell me the difference(s), and a way to ...
5
votes
1answer
628 views

Patent-free Interest Point Detector? [duplicate]

Possible Duplicate: Free alternatives to SIFT/SURF for commercial use There are same very good algorithms for interest point detection/description, like: SIFT (Scale-Invariant Feature Transform) ...
5
votes
1answer
403 views

image feature extraction

We are given a set of grayscale image patches obtained from images after edge detection. Each patch is 10x10 pixels with intensity varying between 0 and 255 for each pixel. This set may contain a very ...
5
votes
2answers
105 views

What makes an image feature stable?

Feature detection is an essential task in low-level vision. Good features are those that resist to different perturbations such as noise addition, blur, geometric transforms (3D rotation with ...
5
votes
0answers
145 views

Filtering Corners Detected on Multiple Scales

I made a multi-scale Harris corner detector (inspired by MOPS) which is simply a Harris corner detector performed on several scales (i.e. subsampled versions of the same image). This effectively make ...
4
votes
2answers
122 views

Useful natural “Hilbert-like” $n$-uples and $n$-fold "analytic signals

If $\mathcal{H}$ denotes the Hilbert transform, the analytic signal of a signal $x(t)$ is $$x_a(t) = x(t) +\imath \mathcal{H}(x(t))\,.$$ The real and imaginary parts form Hilbert pairs. Are there ...
4
votes
3answers
745 views

Local Image Features Extraction for Road Signs

I'm working on a project to detect a road sign from the Image. Consider the sample image below: I'm using a heuristic search to detect the road sign within an image. Basically, I want some weak ...
4
votes
1answer
201 views

What scale should a detected SIFT point be extracted at?

I've heard this from a few different researchers but I can't find a definitive value. Given the "scale" extracted by David Lowe's SIFT detector, we need to extract a descriptor at a specific scale. ...
4
votes
2answers
182 views

mean of wavelet for image processing

For many papers which talks about using wavelet transforms for feature detection in image processing, it is stated that it is advantageous for the wavelet to have zero mean? Why is this so? Thanks in ...
4
votes
1answer
1k views

Structure Tensor vs. Hessian Matrix

Hello can someone explain the semantical meaning of the structure tensor and the Hessian matrix. I am aware of how it is calculated, but i find it difficult to comprehend what they describe in the ...
3
votes
2answers
3k views

Methods to quantify randomness (or complexity) in a signal

What are the methods through which we can quantify the randomness or complexity in a given signal. I know spectral flatness measure (geometric to arithmetic means) is one way to do it, but what are ...
3
votes
3answers
738 views

What are integration scale and differentiation scale?

In scale-adapted Harris detector, the scale adapted second moment matrix is defined by: $$\mu(x, \sigma_I, \sigma_D) = \sigma_D^2\ g(\sigma_I) *\left[ \begin{array}{cc} L_x^2(x, \sigma_D) &L_xL_y(...
3
votes
1answer
175 views

What Is an Oriented Gaussian Second Derivative Filter

In the paper: Detecting and Localizing Edges Composed of Steps, Peaks and Roofs, the authors refer to an image filter as an oriented second-derivative Gaussian filter. I'm trying to figure out what ...
3
votes
3answers
5k views

Principal Component Analysis (PCA) on Convolutional Neural Network (CNN) Features

Please, I have a question regarding PCA and features which are extracted from a convolutional layer. link if we have a test dataset , and we extract all conv features of all images at test dataset ...
3
votes
1answer
697 views

proving that a log gabor filter has 0 DC offset

I have read somewhere online that the log gabor filter has an advantage over the gabor filter, in the sense that it has 0 DC component. How do you prove this property mathematically? Thanks in advance....
3
votes
1answer
1k views

Determining false positives in feature matching

I extracted features from an image using SIFT and SURF. Then I applied few transforms on the image and again calculated the features. While matching the features, I noticed that there are some ...
2
votes
1answer
100 views

What is localizability in Computer Vision?

Please consider the following excerpt (from Rodrigo R, Zouqi M, Chen Z, Samarabandu J.: Robust and efficient feature tracking for indoor navigation): Robust feature tracking is a requirement for ...
2
votes
1answer
7k views

blob detection using Laplacian of Gaussian filter

I'm learning feature-detectors from this lecture notes, and I don't quite understand the Normalized Laplacian of Gaussian ...
2
votes
2answers
3k views

A feature extraction method: Principal Component Analysis (PCA)

I read some articles about PCA and I think the nice way to summarize it is: It is a way of identifying patterns in data, and expressing the data in such a way as to highlight their similarities and ...
2
votes
3answers
238 views

Can I use standard computer vision techniques for images taken in the NIR spectral range?

I'm relatively new in Computer Vision and I need some advice with respect to the general possibilities and shortcomings of images taken in the near infrared (NIR) range. I passed a great deal of time ...
2
votes
2answers
312 views

Detecting sound inside a sound

I was searching on the Internet hopelessly about some materials regarding detecting a sound in another sound. Say, I've got a recorded short sound (which may be anything, fragment of speech, fragment ...
2
votes
1answer
123 views

Feature matching of images without corners

I am trying to stitch aerial images using feature-based registration. Most image pairs are matched well, such as this: Others are not matched due to lack of common features: As you can see, the ...
2
votes
3answers
7k views

normalized Laplacian of Gaussian

Laplacian of Gaussian formula for 2d case is $$\operatorname{LoG}(x,y) = \frac{1}{\pi\sigma^4}\left(\frac{x^2+y^2}{2\sigma^2} - 1\right)e^{-\frac{x^2+y^2}{2\sigma^2}},$$ in scale-space related ...
2
votes
1answer
5k views

Zernike Moments' implementation in OpenCV

The reason why Hu moments were implemented in OpenCV and why Zernike moments were not implemented is looking like their performance similar as stated in this paper. As stated in the paper Zernike ...
2
votes
1answer
107 views

feature detectors and descriptors comparison

There are several kinds of detectors and descriptors, like SIFT, SURF, FAST. I wonder are they all eligible for real-time applications? Which is the best or better? ...
2
votes
1answer
2k views

Understanding Integration Scale in Multi-Scale Harris Corner Detector

I would like to improve a simple Harris corner detector by adding more scales. I don't need scale-invariant interest points, just to detect more points by processing more scales. The Wikipedia says ...
2
votes
0answers
974 views

Relationship between Wavelet transform and Fourier Power Spectral Density

Is there anyway to obtain the Fourier Power Spectral Density from a wavelet transform of a time series? I am particularly interested in this problem because I was wondering if there is any ...
2
votes
0answers
121 views

what is in intuitive explanation of local derivative pattern (LDP)?

Zhang et al. proposed the LDPs for face recognition . They considered the LBP as the nondirectional first-order local pattern operator and extended it to higher orders ( th-order) called the LDP. is ...
2
votes
1answer
300 views

Haralick features, am I approaching this correctly?

I'm trying to personalise a custom CBIR by adding more features not only based on colour, but based on textures' features. What I don't know is if I'm approaching this well with Haralick's features. ...
2
votes
0answers
457 views

Efficient Hessian-Laplace blob detector implementation

As it is mentioned in a paper for SURF, it is possible to approximate hessian determinant using integral images. If I want to implement Hessian-Laplace detector, is it feasible to also approximate ...
2
votes
0answers
3k views

Why is LBP generaly faster than HAAR?

I am digging into Haar like and LBP features. More precisely its implementation in OpenCV. Every article or forum entry I found states, that LBP is faster than Haar. My local tests also confirm this ...
2
votes
0answers
2k views

MPEG-7 descriptors' implementation in OpenCV

I'm looking for implementation of MPEG-7 descriptors which are compatible with OpenCV's recent versions and containing "region-based shape descriptor ART(Angular Radial Transformation)". I ...
2
votes
1answer
231 views

Can SIFT run in realtime?

What is the best possible runtime for extracting SIFT keypoints and Descriptors? I know it depends on #keypoints extracted. So, say for image of size 640x640, the code that I have been using requires ...
2
votes
0answers
177 views

Affine detector + SIFT descriptor

I am approaching to the Oxford Buildings dataset which contains already computed SIFT descriptors from affine detected keypoint locations. Usually SIFT detector provides to its detected keypoints ...
1
vote
2answers
975 views

texture matching between patches of an image

I have an image A which I have divided into 4 x 4 subband images. For a given patch P1 in ...
1
vote
3answers
231 views

Need critical help: How to detect and distinguish two very similar looking signals?

Hi guys I have a really tough signal processing question here. How do you detect and distinguish two very very similar looking waves? I need to distinguish between these two signals for an online ...