Image Processing Method Using Low-Pixel Borderline Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image processing methods are inefficient in identifying target images from original high-resolution images due to the time-consuming process of scanning pixels, resulting in long acquisition times and low efficiency.

Innovation Solution

The method involves converting the original image into a low-pixel image using a preset compression ratio, determining borderlines in the low-pixel image, and mapping these borderlines back into the original image to acquire the target image, utilizing a fast bilinear interpolation algorithm and edge detection techniques.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If pixels in the original image are scanned one by one to determine borderlines, then the target image can be accurately identified, but the processing time becomes very long and efficiency is low

Engineering Contradiction:
Improvetarget image identification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the image processing task into two segments: first processing a down-sampled low-pixel image to identify target borderlines, then mapping these borderlines to the original high-resolution image. This segmentation allows the computationally intensive borderline detection to be performed on fewer pixels while still achieving accurate target identification in the original image.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a down-sampled copy of the original image with reduced pixel density. This copy is used for preliminary target detection and borderline identification. The identified borderlines from the copy are then transferred back to the original image, avoiding the need to process all pixels in the original image directly.

Inventive Principle:
Principle #26Copying

2Manufacturing precision

If the original high-resolution image is processed directly, then the target image quality is maintained, but the number of pixel points increases processing complexity and time

Engineering Contradiction:
Improvetarget image qualityVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent introduces a dimensionality reduction step by creating a down-sampled version of the image with fewer pixels. This reduces the computational dimensionality of the processing task. The solution then maps the results back to the original dimensional space, effectively solving the problem in a lower dimension and transferring the solution to the full dimension.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent performs preliminary processing on the down-sampled image to identify target borderlines before applying these results to the original high-resolution image. This preliminary action on the simplified version prepares the solution that is then transferred to the full-resolution image, avoiding direct complex processing of all original pixels.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3306562B1Image processing method and device
Publication Date: 2020.04.08 ALIBABA GROUP HOLDING LTD
  • EP3306562B1 patent drawingFigure 1
  • EP3306562B1 patent drawingFigure 2
  • EP3306562B1 patent drawingFigure 3

AI summary

The present application discloses an image processing method and apparatus. The method includes: converting an original image into a low-pixel image according to a preset compression ratio; performing borderline detection on the low-pixel image to determine borderlines of a target image included in the low-pixel image; and mapping the borderlines included in the low-pixel image into the original image, to acquire the target image included in the original image. By means of the technical solution of the present application, the original image is converted into a low-pixel image. Since the low-pixel image includes fewer pixel points, and the target image is acquired based on the low-pixel image, the target image acquisition process based on the low-pixel image shortens the duration for acquiring the target image, and increases the efficiency of acquiring the target image.