Edge Diversity Object Detection in Environmental Interference

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for detecting objects in images fail to distinguish between objects of interest and their surroundings, particularly in environments with environmental effects like fog, moisture, smog, smoke, and solar glint, leading to false detection due to strong edge intensities from unwanted features.

Innovation Solution

A method that assigns each pixel to orientation bins based on its orientation value, determines the maximum pixel magnitude values for each bin in a surrounding region, sums these values, and generates an updated image with thresholds to identify object regions, while aggregating the image using functions of ground sample distance and solar glint angle to reduce unwanted features.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional edge detection methods are used, then edges from objects of interest can be detected, but edges from unwanted features (surroundings, environmental effects) are also detected and often appear stronger, leading to false positives

Engineering Contradiction:
Improveobject detection accuracyVSAvoidfalse detection rate
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent segments edge detection by orientation, dividing the edge detection space into multiple orientation bins (e.g., 0-45°, 45-90°, etc.). Instead of treating all edges equally, it processes different orientation ranges separately, allowing the system to distinguish between edges of objects of interest and edges from unwanted features based on their directional characteristics

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by analyzing the distribution of edge orientations at each pixel location. It calculates the ratio of edges in different orientation bins within a local neighborhood, creating a signature that is unique to each location. This allows the system to identify whether a pixel belongs to an object of interest or to unwanted features based on the local orientation distribution pattern

Inventive Principle:
Principle #3Local quality

2Measurement precision

If image intensity levels are used for detection, then objects can be identified, but environmental effects (solar glint, fog, moisture, smog, smoke, exhaust plumes) create similar intensity patterns that confuse the detection system

Engineering Contradiction:
Improveobject detection accuracyVSAvoidenvironmental interference
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent adds the orientation dimension to traditional intensity-based detection. Instead of relying solely on intensity magnitude, it incorporates the orientation of edges as a separate dimension of analysis. This dimensional expansion allows the system to distinguish between intensity patterns caused by objects of interest and those caused by environmental effects, as they typically have different orientation characteristics

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

Solution Approach 2:

The patent changes the detection parameters from simple intensity magnitude to a combination of orientation angle and intensity. By transforming the detection space to include orientation information, the system can identify objects based on their characteristic edge orientations rather than being fooled by intensity similarities to environmental features

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2453409B1Edge diversity object detection
Publication Date: 2016.09.07 RAYTHEON CO
  • EP2453409B1 patent drawingFigure 1
  • EP2453409B1 patent drawingFigure 2A~2B
  • EP2453409B1 patent drawingFigure 2C~2D

AI summary

Methods for detecting objects in an image. The method includes a) receiving magnitude and orientation values for each pixel in an image and b) assigning each pixel to one of a predetermined number of orientation bins based on the orientation value of each pixel. The method also includes c) determining, for a first pixel, a maximum of all the pixel magnitude values for each orientation bin in a predetermined region surrounding the first pixel. The method also includes d) summing the maximum pixel magnitude values for each of the orientation bins in the predetermined region surrounding the first pixel, e) assigning the sum to the first pixel and f) repeating steps c), d) and e) for all the pixels in the image.