Questions tagged [computer-vision]

Please utilize this tag when inquiring about Computer Vision, which encompasses all software aspects that allow computers to perceive, comprehend, and respond to their surroundings through the use of cameras. If your questions pertain to image filtering and quantification, please opt for the [image-processing] tag instead.

Convolutional layers featuring an image with three channels

I am looking to implement the DQN code from the Keras-rl library (https://github.com/matthiasplappert/keras-rl/blob/master/examples/dqn_atari.py) for 3-channel images without converting them to grayscale. How can I modify the code to achieve this? I attem ...

Utilizing Scikit-image for extracting features from images

I have been using scikit-image to successfully classify road features. Take a look at the results here: https://i.stack.imgur.com/zRJNh.jpg. However, I am facing challenges in the next step of classifying these features. Specifically, I need to classify fe ...

Python code: Transforming points on a contour into a bounding box

Currently, I am analyzing medical images using XML files that contain contour coordinates detailing regions of interest. Despite successfully extracting these points, I am facing challenges in converting them into a bounding box suitable for creating masks ...

Converting the OpenCV GetPerspectiveTransform Matrix to a CSS Matrix: A Step-by-Step Guide

In my Python Open-CV project, I am using a matrix obtained from the library: M = cv2.getPerspectiveTransform(source_points, points) However, this matrix is quite different from the CSS Transform Matrix. Even though they have similar shapes, it seems that ...

Create hough transformation lines on top of a photo

Trying to implement the Hough line transform independently has proved challenging, particularly when it comes to drawing the high-voted thetas/rhos values. Despite attempting the mathematical calculations myself, the outputs were consistently inaccurate. U ...

Discovering the conversion of the camera matrix

Here is another inquiry regarding computer vision. The concept of a camera matrix, also known as a projection matrix, involves the mapping of a 3D point X from the real world to an image point x in a photograph using the equation: l **x** = P **X** Esse ...

What are the steps to implement OpenCV for a real-time application?

When it comes to real time systems, a quick and efficient platform is essential. I'm curious about which combination would be best for real time image processing: OpenCV-Python, OpenCV-MVStudio (C++), OpenCV-Matlab, or OpenCV-Java? I've heard abo ...

Looking for a way to locate a reference point within an object for measurement using OpenCV in Python?

Click here to upload an image for reference point identification The task at hand involves measuring the length of a drill bit. If two reference points can be automatically identified, the Euclidean distance formula can then be applied to measure the dist ...

What is the reason for OpenCV assuming that I am utilizing a calibration rig that is not planar?

Exploring the functionality of OpenCV's calib3d module, I am currently experimenting with a simulated camera rig. To better grasp how it works and performs, I have generated a set of artificial object points in 3D space that correspond to a flat grid ...