Image Recognition Apparatus Region Rotation for Ceiling Camera Inversion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Image recognition apparatuses struggle to detect objects in images captured by ceiling cameras, where the positional relationship between the object's head and feet is inverted compared to learned data, preventing the use of preexisting position data.

Innovation Solution

The apparatus divides the image into regions, sets a process region, and rotates it so that the positional relationship between the object's head and feet matches the standard orientation, allowing for recognition using preexisting dictionaries or deep models.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image recognition uses preexisting position data learned from standard images, then recognition accuracy is improved, but the system cannot handle images with inverted positional relationships (ceiling camera views)

Engineering Contradiction:

Improverecognition accuracyVSAvoidadaptability to different camera orientations

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The image is divided into multiple regions based on a positioning point, allowing different parts of the image to be processed independently. This segmentation enables the system to handle inverted positional relationships in specific regions while maintaining standard processing for others.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies rotation to process regions to invert the positional relationship when necessary. By rotating regions 180 degrees, the system transforms ceiling camera views with inverted head-feet relationships into standard orientations that match preexisting recognition models.

Inventive Principle:

Principle #13The other way round (Inversion)

2Adaptability or versatility

If the system rotates process regions to match standard orientation, then compatibility with preexisting recognition models is improved, but processing complexity increases

Engineering Contradiction:

Improvecompatibility with preexisting modelsVSAvoidimage processing complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

Different processing treatments are applied to different regions of the image. Only specific process regions requiring orientation correction are rotated, while other regions maintain their original orientation. This localized approach reduces overall processing complexity compared to rotating the entire image.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system performs preliminary determination of which regions require rotation before executing the recognition process. By identifying and marking process regions that need orientation correction in advance, the system optimizes the processing workflow and avoids unnecessary rotations.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11475541B2Image recognition apparatus

Publication Date: 2022.10.18 KK TOSHIBA

AI summary

An image recognition apparatus includes circuitry. The circuitry is configured to input an image of an object captured by a camera. The circuitry is further configured to divide, based on a predetermined positioning point, the image into a plurality of regions, set a process region that includes the respective divided region, and set a rotation of the respective process region so that a positional relationship between up and down of the object in the respective process region matches up. The circuitry is further configured to perform the rotation to the image corresponding to the respective process region and perform a recognition process to the image after rotation.