Camera Frame Segmentation for Small Target Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video monitoring systems are inefficient in detecting and positioning small targets within images due to downscaling issues that can render small objects unrecognizable, and they often require additional dedicated devices or modifications to the objects being tracked.

Innovation Solution

A monitoring system that segments camera frames into smaller segments with resolutions matching the reference images, allowing for precise visual search operations without losing details, using modules for segmentation, visual search, position calculation, and consistency checks to determine the position of small targets within the monitored area.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the image is downscaled to match reference image resolution, then the visual search operation can be performed efficiently, but small targets become unrecognizable and detection precision deteriorates

Engineering Contradiction:
Improvevisual search efficiencyVSAvoidtarget detection precision
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent divides the high-resolution image into multiple segments, each segment being processed separately at the reference image resolution. This segmentation allows the visual search to operate efficiently on smaller regions while maintaining the ability to detect small targets that would be lost in a full-image downsacle.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If additional dedicated devices or object modifications are used to improve small target detection, then detection precision improves, but device complexity and cost increase

Engineering Contradiction:
Improvesmall target detection precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes the existing camera system self-sufficient by using software-based image processing techniques (segmentation and coordinate transformation) rather than requiring additional hardware devices or modifications to the objects being monitored. The system uses its own resources to achieve improved small target detection.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If the full-resolution image is processed directly, then target detection precision is maintained, but processing time and computational resources increase significantly

Engineering Contradiction:
Improvetarget detection precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the high-resolution image into multiple smaller regions, processing each segment independently at the reference image resolution. This reduces the computational burden compared to processing the entire high-resolution image, while still maintaining detection precision through coordinate transformation of results back to the original image scale.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies visual search operations to multiple segments rather than the entire image at once. Each segment is processed with partial action (at reference resolution), and the results are aggregated to achieve complete coverage, balancing processing efficiency with detection completeness.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS11003963B2Method and system for identifying targets in scenes shot by a camera
Publication Date: 2021.05.11 TELECOM ITALIA SPA
  • US11003963B2 patent drawing
  • US11003963B2 patent drawing
  • US11003963B2 patent drawing

AI summary

A monitoring system assesses the position of at least one target within an area. The monitoring system comprises a processing unit adapted to receive, from a camera, at least a frame depicting a scene of the area. The frames generated by the camera have a first resolution. The processing unit comprises: a database module storing, for each at least one target, a corresponding reference image, each reference image having a second resolution lower than the first resolution; a segmentation module configured to segment a received frame into frame segments, each frame segment having a resolution corresponding to the second resolution; a search module configured to implement a visual search procedure between each frame segment and a reference image comprising a depiction of the at least one target; and a position and translation module for calculating the position of the at least one target within the received frame.