Image Tracking Using Distance and Pixel Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing technologies face challenges in tracking moving objects with the same color or luminance, leading to potential misidentification and loss of tracking focus when multiple objects intersect, as they rely solely on color or luminance information.

Innovation Solution

An image processing apparatus that utilizes both distance information and pixel value information to track moving objects, allowing for precise identification and tracking by distinguishing between objects based on their distance from the camera and color/luminance values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If tracking is performed based solely on color or luminance information, then the tracking process is simple, but tracking precision deteriorates when multiple objects have the same color or luminance

Engineering Contradiction:
Improvetracking process complexityVSAvoidtracking precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces distance information as an additional dimension to the tracking process. Instead of relying solely on color or luminance (2D color space), the system now uses a combination of color/luminance and distance (adding the depth dimension), creating a 5-dimensional tracking parameter space (x, y, color, luminance, distance). This dimensional expansion allows differentiation of objects that would otherwise be indistinguishable in the original color-luminance space.

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

Solution Approach 2:

The patent changes the tracking parameters from just color and luminance to include distance information. By modifying the set of parameters used for object identification and tracking, the system can distinguish between multiple objects with identical color and luminance values. The distance parameter acts as an additional discriminator that resolves ambiguities in object identification.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If distance information is acquired and used for tracking, then tracking precision improves, but device complexity increases

Engineering Contradiction:
Improvetracking precisionVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes the image sensor perform multiple functions: it serves both as the primary imaging device for capturing visual information and as a distance measurement device through focus detection. By utilizing the phase difference information already present in the image sensor output, the system avoids adding separate distance measurement hardware, thereby reducing overall device complexity while still achieving precise tracking through multi-parameter identification.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent combines the imaging function and distance measurement function into a single integrated system. The image sensor and its focus detection capability are merged with the tracking system, allowing simultaneous acquisition of visual information and distance data. This merging eliminates the need for separate components and reduces system complexity while improving tracking precision through the use of combined parameters.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP2793457B1Image processing device, image processing method, and recording medium
Publication Date: 2019.05.08 SONY GROUP CORP
  • EP2793457B1 patent drawingFigure 1
  • EP2793457B1 patent drawingFigure 2
  • EP2793457B1 patent drawingFigure 3

AI summary

Provided is an image processing apparatus including a distance information acquisition unit that acquires distance information on a distance up to an object imaged by an image sensor, a pixel value information acquisition unit that acquires pixel value information of an image corresponding to the object, and a tracking unit that tracks the object that moves, based on the acquired distance information and the acquired pixel value information.