A carpet recognition method based on LED light source

By using an LED light source-based carpet recognition method and grayscale image processing and discrimination formulas, the problem of environmental influence when a robot vacuum cleaner recognizes carpets is solved, achieving high-precision and low-cost recognition results, which is suitable for small electrical appliances.

CN116310269BActive Publication Date: 2026-06-05SHENZHEN YOUXIANG COMPUTING TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHENZHEN YOUXIANG COMPUTING TECH CO LTD
Filing Date
2023-01-08
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

When existing robotic vacuum cleaners use ultrasonic sensors to identify carpets, they are easily affected by ambient temperature and noise, resulting in a low carpet recognition rate.

Method used

A carpet recognition method based on LED light source is adopted. By obtaining the center point and radius of the light spot in the grayscale image, the concentric image region is divided, the grayscale mean and variance are calculated, and the carpet is recognized twice using the discriminant formula. Accurate recognition is achieved by combining the grayscale histogram and gradient change feature value.

Benefits of technology

It improves the accuracy and reliability of carpet recognition, reduces equipment requirements, is suitable for small electrical appliances, lowers application costs, and meets real-time processing needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116310269B_ABST
    Figure CN116310269B_ABST
Patent Text Reader

Abstract

The application relates to a carpet recognition method based on an LED light source, comprising the following steps: S1. photographing a ground irradiated by an LED light source and obtaining a corresponding gray-scale image; S2. obtaining a center point pixel (x1, y1) of a light spot in the gray-scale image; S3. taking the center point pixel (x1, y1) of the light spot as the center, a light spot radius about the light spot is obtained; S4. taking the center point pixel (x1, y1) as the center, a plurality of concentric image regions are divided according to the light spot radius as a unit length, and the gray-scale mean and the gray-scale variance of each image region are calculated respectively; S5. based on the gray-scale mean and the gray-scale variance of the plurality of image regions, first carpet recognition is carried out, if the first recognition result is carpet, second carpet recognition is carried out on the gray-scale image, and if the second recognition result is still carpet, it is determined that the carpet is carpet.
Need to check novelty before this filing date? Find Prior Art