3D Imaging Method Using Time-of-Flight Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 3D active imaging systems by cutting into successive planes require significant memory and processing resources and are dependent on video frame rate, limiting their ability to provide real-time 3D images efficiently.

Innovation Solution

A method utilizing a device with a matrix of optoelectronic sensors that emits light pulses at varying time-of-flight, captures reflected signals, integrates them based on duration, compares with a threshold, and stores color data in a buffer memory, allowing for direct image transfer without relying on video frequency, optimizing acquisition time by varying the number of pulses per sub-process based on depth and executing sub-processes in decreasing order of depth.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all images of successive planes are stored entirely in memory and then added together to produce the 3D image, then the 3D image can be constructed, but the memory requirements and processing means increase significantly

Engineering Contradiction:
Improve3D image construction accuracyVSAvoidmemory requirements
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent divides the scene into multiple depth planes and processes each plane separately through successive light pulse emissions at different time-of-flight values. Instead of storing all plane images in memory, the system processes and displays each plane's contribution sequentially, segmenting the memory requirement into manageable portions corresponding to individual depth planes rather than requiring simultaneous storage of all planes.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary processing of each depth plane's image data as it is captured, immediately integrating it into the final 3D image representation. This preliminary action eliminates the need to store complete plane images in memory for later processing, as the integration happens in real-time during the acquisition phase itself.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If all images of successive planes are stored entirely in memory and then added together to produce the 3D image, then the 3D image can be constructed, but the processing means increase significantly

Engineering Contradiction:
Improve3D image construction accuracyVSAvoidprocessing means
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the processing task into separate operations for each depth plane, where each plane's data is processed independently and immediately integrated into the final image. This segmentation allows for simpler, more efficient processing circuitry that handles one plane at a time rather than requiring complex simultaneous processing of all planes.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements continuous processing where each depth plane's image data is integrated into the final 3D image as it is captured, without interruption or intermediate storage. This continuous integration process eliminates the need for separate post-processing steps, reducing overall processing complexity while maintaining image construction accuracy.

Inventive Principle:
Principle #20Continuity of useful action

3Ease of operation

If the system depends on video frame rate to obtain the 3D image, then the image can be displayed, but real-time performance is limited

Engineering Contradiction:
Improveimage display capabilityVSAvoidreal-time imaging speed
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent uses periodic emission of light pulses at different time-of-flight values to sequentially probe different depth planes. This periodic action allows the system to acquire depth information for multiple planes in rapid succession, with each pulse contributing to the final 3D image without being constrained by traditional video frame rates, thereby achieving real-time performance.

Inventive Principle:
Principle #19Periodic action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach reduces memory requirements, minimizes processing needs, and enables real-time 3D image generation independent of video frame rate, with each pixel corresponding to a specific depth, enhancing the efficiency and accuracy of 3D imaging.

Implementation Method 1

a matrix of Cn optoelectronic sensors each able to generate a signal representative of the intensity of the reflected light pulse that it has received

Methodology Applied
Scientific EffectPhotoelectric Effect: Photoelectric Effect

Implementation Method 2

Each pulse captured gives the reflected areas of the scene at a determined time of flight

Methodology Applied
Scientific EffectTime of flight: Time of Flight

Data Source

PatentEP2735886B13D imaging method
Publication Date: 2017.08.30 INSTITUT FRANCO ALLEMAND DE RES & DEVS DE SAINT LOUIS
  • EP2735886B1 patent drawingFigure 1
  • EP2735886B1 patent drawingFigure 2
  • EP2735886B1 patent drawingFigure 3

AI summary

A 3D imaging method decomposed into f sub-processes comprising the following steps: - emitting a light pulse towards a scene, - capturing the light pulse reflected by the scene at a determined time of flight with an array of optoelectronic sensors; - integrating each signal generated by the array; - comparing the level of the integrated signal to a threshold value; - storing in the buffer, at addresses associated with the sensors that captured and integrated the reflected light pulse with a value greater than said threshold value, a color associated with the time of flight of the captured light pulse; - blocking the corresponding addresses; a method comprising the following steps: - executing the sub-processes one after the other; - changing the time of flight and the associated color for each sub-process; - transferring the data from the buffer into the image memory after the last sub-process.