Overlapping Image Sensor Array for High-Resolution Scene Capture

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current machine vision systems are costly, require significant installation space, and are not economically viable for recording large image scenes with high scene resolution and high frame rates without distortion or parallax, especially in small industrial settings where resources are limited.

Innovation Solution

An image sensor arrangement with overlapping detection areas of adjacent sensor cells, each equipped with its own optics and evaluation unit, allows for scalable computing power and resolution, enabling efficient data processing and integration into compact systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional machine vision systems use multiple cameras with separate evaluation units connected via ribbon cables to a central processing system, then image data can be captured and processed, but the system requires significant installation space, has high costs, and complex data transmission infrastructure

Engineering Contradiction:
Improvescene resolutionVSAvoidinstallation space
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

The patent combines multiple sensor cells with overlapping detection areas into a unified sensor array where adjacent cells share data through side connections. This merging eliminates the need for separate evaluation units and ribbon cable connections for each camera, reducing installation space while maintaining high scene resolution through coordinated data processing from multiple sensors.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent transitions from a one-dimensional array of sensors to a two-dimensional sensor array with overlapping detection areas. This dimensional expansion allows adjacent sensor cells to share data horizontally and vertically through side connections, enabling comprehensive scene coverage and high resolution without requiring proportional increases in installation space.

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

2Measurement precision

If a large number of cameras are arranged in a matrix with overlapping detection areas and connected via ribbon cables to a central processing system, then high scene resolution can be achieved, but the data transmission infrastructure becomes complex and costly

Engineering Contradiction:
Improvescene resolutionVSAvoiddata transmission infrastructure
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the data processing function by enabling each sensor cell to independently process its own detection data and share it with adjacent cells through side connections. This segmentation eliminates the need for a complex centralized data transmission infrastructure with ribbon cables, as each sensor unit becomes self-sufficient while contributing to the overall high-resolution scene through local data exchange.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each sensor cell in the patent performs self-service by capturing, processing, and sharing its own detection data with adjacent cells through direct side connections. This self-service capability eliminates the need for external ribbon cable connections and centralized processing infrastructure, reducing device complexity while maintaining the ability to generate high-resolution scene data through coordinated operation of multiple autonomous sensor units.

Inventive Principle:
Principle #25Self-service

3Productivity

If conventional systems use separate evaluation units for each camera connected to a central processing system, then image data can be processed, but the system requires significant installation space and has high costs

Engineering Contradiction:
Improveframe rateVSAvoidinstallation space
Core Design Contradiction:
ProductivityVSArea of stationary object

Solution Approach 1:

The patent merges the evaluation functions of multiple separate cameras into a unified processing approach where adjacent sensor cells share data through side connections. This consolidation eliminates redundant evaluation units and their associated installation space requirements while maintaining high frame rates through efficient local data processing and sharing across the sensor array.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent utilizes a two-dimensional array configuration where sensor cells are arranged in rows and columns with overlapping detection areas. This spatial arrangement enables efficient data sharing between adjacent cells in multiple directions, allowing the system to achieve high processing throughput and frame rates without proportionally increasing installation space, as the evaluation function is distributed across the two-dimensional sensor landscape rather than requiring separate vertical stacking of processing units.

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

Data Source

PatentEP2491713B8Image sensor array for detecting image information for automatic image data processing
Publication Date: 2018.05.23 XAPT GMBH

AI summary

The invention relates to an image sensor array (1) and to a method for detecting image information for automatic image data processing. The image sensor array (1) comprises a plurality of identical sensor cells (2), each being provided with an image sensor (3) having a dedicated sensor optics (8), an image store for storing image information of the image sensor (3), and an evaluation unit (4) for processing the image information. The sensor cells (2) are arranged in at least one series such that the detection ranges (7) of the image sensors (3) of directly adjacent sensor cells (2) overlap. Each sensor cell (2) comprises at least one data interface (5) to an adjacent sensor cell (2), by means of which the image data of the overlapping detection range (7) of the adjacent sensor cell (2) can be accessed. The evaluation units (4) are equipped to determine a correlation of said image data.