Target Detection for Visually Impaired Using Alignment Prompts

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In image processing for visually impaired individuals, misalignment of the image sensor with the target object can lead to recognition issues, particularly when text is involved, due to the lack of a preview image or visible display, causing difficulties in accurately detecting and recognizing text or objects.

Innovation Solution

A target detection method that includes acquiring an image, detecting the presence of a target object using bounding box detection, and providing prompts to adjust the object or image sensor to ensure proper alignment, thereby improving image quality and recognition accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If bounding box detection is performed to detect target objects, then target detection capability is improved, but recognition accuracy deteriorates when text is involved due to misalignment

Engineering Contradiction:
Improvetarget detection capabilityVSAvoidrecognition accuracy
Core Design Contradiction:
Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The system performs preliminary alignment detection by checking whether the bounding box is properly aligned with the image boundaries before attempting text recognition. If misalignment is detected, the system prompts the user to adjust the image capture position beforehand, preventing recognition errors caused by misalignment

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system provides feedback to the user by displaying prompts that indicate whether the bounding box is properly aligned with the image boundaries. This feedback loop allows users to adjust their capture position based on real-time alignment information, improving recognition accuracy

Inventive Principle:
Principle #23Feedback

2Device complexity

If image processing is performed without preview display, then device complexity is reduced, but operation ease deteriorates for visually impaired users

Engineering Contradiction:
Improvedevice complexityVSAvoidoperation ease
Core Design Contradiction:
Device complexityVSEase of operation

Solution Approach 1:

The system replaces visual preview display with auditory feedback mechanisms. Instead of showing a preview image on a display screen, the system uses audio prompts to communicate alignment status and processing results to visually impaired users, maintaining operational ease without requiring visual display hardware

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system serves multiple functions through a single audio output channel: providing alignment guidance, indicating processing status, and communicating detection results. This multi-functionality eliminates the need for separate visual display while maintaining comprehensive user feedback

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

3Measurement precision

If alignment adjustment prompts are provided, then recognition accuracy is improved, but processing time increases

Engineering Contradiction:
Improverecognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs alignment detection and provides prompts before the actual text recognition processing begins. By ensuring proper alignment in advance through preliminary checks and user guidance, the system prevents time-consuming reprocessing due to misalignment, actually reducing overall processing time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system automatically detects alignment status and generates appropriate prompts without requiring manual intervention or extended processing. The alignment detection operates autonomously, and the prompts are generated instantaneously, minimizing added processing time while ensuring accuracy

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3842992B1Target detection method, circuit, device for assisting visually impaired people, electronic device, and medium
Publication Date: 2023.06.21 NEXTVPU (SHANGHAI) CO LTD
  • EP3842992B1 patent drawingFigure 1~2B
  • EP3842992B1 patent drawingFigure 3~4
  • EP3842992B1 patent drawingFigure 5A~5B

AI summary

Provided are a target detection method, a circuit, a device for assisting visually impaired people, an electronic device, and a medium. The target detection method includes: acquiring an image collected by an image sensor; detecting whether there is at least a portion of a target object in the image; and providing, in response to not detecting at least a portion of the target object in the image, a prompt to adjust the target object to be within a field of view of the image sensor.