A fast image recognition algorithm for box?

Click For Summary
SUMMARY

The discussion focuses on identifying the four corners of a box in a noisy black and white image using image processing techniques. The recommended approach involves applying a 2D low pass filter, specifically a Gaussian filter, to reduce noise before detecting edges. The process includes convolving the image with a Gaussian matrix or utilizing the Fast Fourier Transform (FFT) to enhance the image quality. Following the filtering, calculating the gradient in both the x and y axes helps pinpoint where the box appears in the image.

PREREQUISITES
  • Understanding of 2D low pass filtering techniques
  • Familiarity with Gaussian blur and its application
  • Knowledge of Fast Fourier Transform (FFT) in image processing
  • Basic concepts of image gradients and edge detection
NEXT STEPS
  • Research Gaussian blur implementation in Python using OpenCV
  • Learn about Fast Fourier Transform (FFT) for image processing
  • Explore edge detection techniques using gradients in images
  • Investigate other noise reduction algorithms suitable for image preprocessing
USEFUL FOR

Image processing engineers, computer vision researchers, and developers working on applications requiring robust box detection in noisy environments.

nhmllr
Messages
183
Reaction score
1
If you have an image (just black and white) with a box SOMEWHERE in the image, but there's also some noise in the image, is there a good way to find four corners?

I'm sure someone has herd of an algorithm for this, it would just be good if I had a direction to go in.
 
Technology news on Phys.org
nhmllr said:
If you have an image (just black and white) with a box SOMEWHERE in the image, but there's also some noise in the image, is there a good way to find four corners?

I'm sure someone has herd of an algorithm for this, it would just be good if I had a direction to go in.

I would advise to use a 2D low pass filter first to get rid of the noise as much as possible. In other means convolve the image with a gaussian matrix. Or in other saying take fft of the image and multiply with the fft of the gaussian matrix. And then take the inverse of the solution. Dimensions of the gaussian matrix will determine the cut of spatial frequency:

http://en.wikipedia.org/wiki/Gaussian_blur#Sample_Gaussian_matrix

There should be a builtin gaussian filter in the environment you are working. Other than that if I were you I would take a gradient of the filtered image both x and y-axis to see in which pixels the box started to appear.
 
Last edited:

Similar threads

Non youtube courses and paid textbooks to learn algorithms?
  • · Replies 8 ·
Replies
8
Views
4K
What is the fastest algorithm for finding large prime numbers?
  • · Replies 14 ·
Replies
14
Views
3K
Fast reciprocal square root algorithm
  • · Replies 30 ·
2
Replies
30
Views
7K
Challenge: Find the minimum amount of draws possible for a list of integers
  • · Replies 9 ·
Replies
9
Views
3K
How Can We Efficiently Parallelize the N-Queens Problem?
  • · Replies 11 ·
Replies
11
Views
5K
Engineering Do Identical Histograms Remain the Same After Applying a Smoothing Mask?
  • · Replies 2 ·
Replies
2
Views
5K
Final momentum of box with reflective interior hit by ideal laser
  • · Replies 4 ·
Replies
4
Views
1K
Are there standard axes/orientations for Rubik's Cube game programs?
  • · Replies 14 ·
Replies
14
Views
3K
Help with 3-D interactive QM visualisation of a hydrogen atom
  • · Replies 5 ·
Replies
5
Views
2K
Stargazing Astro Image Stacking (optimize DSS image stacking and post processing)
  • · Replies 43 ·
2
Replies
43
Views
12K