Rotation Invariant Object Feature Recognition in Low-Power Cameras

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Cameras and monitoring devices with limited processing power face challenges in object and feature recognition, as they often require image data to be compressed or transmitted to more powerful devices for processing, leading to delayed recognition and inefficient use of bandwidth.

Innovation Solution

A method that enables cameras to detect or recognize features in images using image data before compression, by determining the average intensity of blocks within the image and generating characteristic numbers based on these comparisons, allowing for improved feature quantification and reduced processing requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image data is compressed or transmitted to more powerful devices for processing, then object recognition accuracy is improved, but processing speed and bandwidth efficiency deteriorate

Engineering Contradiction:
Improveobject recognition accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The image is divided into multiple blocks, and feature extraction is performed on each block independently. This segmentation allows the camera's limited processor to handle smaller, more manageable units of data, enabling real-time feature extraction without requiring full image transmission to external devices.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the essential feature information (average intensity values and characteristic numbers) from the image data at the camera端, rather than transmitting the entire compressed image. This extraction of critical features reduces the data transmission requirement and allows the camera to perform recognition-based processing independently.

Inventive Principle:
Principle #2Taking out (Extraction)

2Difficulty of detecting and measuring

If image data is transmitted to more powerful devices for processing, then feature detection capability is improved, but bandwidth utilization deteriorates

Engineering Contradiction:
Improvefeature detection capabilityVSAvoidbandwidth utilization
Core Design Contradiction:
Difficulty of detecting and measuringVSLoss of energy

Solution Approach 1:

The patent extracts only the essential feature information (average intensity values and characteristic numbers) from the image data at the camera端, rather than transmitting the entire compressed image. This extraction of critical features reduces the data transmission requirement and allows the camera to perform recognition-based processing independently.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The camera's limited processor is empowered to perform feature extraction and object recognition independently without requiring external processing assistance. The system uses the camera's own computational resources to calculate block intensities and generate characteristic numbers, making the camera self-sufficient for basic recognition tasks.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3179410B1Rotation invariant object feature recognition
Publication Date: 2019.02.20 AXIS
  • EP3179410B1 patent drawingFigure 1A~1B
  • EP3179410B1 patent drawingFigure 2~3
  • EP3179410B1 patent drawingFigure 4A

AI summary

A method may include determining a value indicative of an average intensity of blocks in an image. The blocks include a primary and outer blocks. Each of the outer blocks may have three, five, or more than five pixels. The image may describe an external pixel lying between the primary and at least one of the outer blocks. The external pixel may not contribute to the value indicative of the average intensity of any of the blocks. The image may also describe a common internal pixel lying within two of the blocks. The common pixel may contribute to the value indicative of the average intensity of the two of the blocks. The method may include comparing the value indicative of the average intensity of the primary block to the values of the outer blocks, and quantifying a feature represented by the image by generating a characteristic number.