Code Monkey home page Code Monkey logo

autonomous-driving-vision's Introduction

장애물 탐지를 위한 이미지 분석 (Obstacle detection)

개발 일지

ezgif com-gif-maker (5)

ezgif com-video-to-gif (11)

We will use object detection, image segmentation, and depth images to make judgments to avoid objects.

1. Receive image and point cloud from camera.

2. Use object detection, image segmentation, depth images to determine obstacles.

  • Obstacle detection begins with the assumption that the portion of the segmentation that is not recognized as a sidewalk is an obstacle.

  • However, considering errors in segmentation, it is more efficient to detect obstacles by object detection. Object detection was designated as the first priority.

  • The y-value of the point cloud was used to limit the effective range of obstacles.

3. We need to know which direction these obstacles are located.

  • Calculate the angle corresponding to each pixel with the x and z coordinates of the point cloud.

4. It is now time to distinguish the types of obstacles.

  • Generally, for mobile devices, avoiding preparations should be made from a distance.

  • however, For fixed things, a closer distance can be accommodated.

image

  • This part can be modified fluidly.

5. Determine the path

ezgif com-video-to-gif (10)

  • The number of pixels per angle was calculated for obstacles detected within 25 meters.

  • And averaged the number of obstacle pixels between -10 and 10 degrees from each angle.

  • The angle was selected by ordering the number of obstacle pixels low and close to 0 degrees.

  • Among them, the angle with the least obstacles was selected.

The test environment is RTX2060 super (fps 15)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.