IMAGE PROCESSOR AND IMAGE PROCESSING METHOD

Patent №

US 8,107,725

Granted

2012-01-31

Filed 2008

Owner

TOYOTA JIDOSHA KABUSHIKI KAISHA

Lab

AI components

1

vision

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

12325444

An image processor according to the present invention is configured to generate a plurality of multiresolution images of different resolutions from an input image (S16), to set a correlation evaluation function for each multiresolution image, calculate a correlation value between the correlation evaluation function and each pixel in the multiresolution image, and extract a position of a local region on the basis of the correlation value (S18), and to set a size of the local region according to the resolution of the multiresolution image (S22), and to detect an object in the local region. This enables extraction of the local region at an appropriate position and in an appropriate range for a characteristic portion of the input image. For this reason, a target range is appropriately limited and detection of the object is quickly carried out without deterioration of accuracy.

VisionG06V 10/56G06V 10/462

AI classification

Vision1.00
Knowledge representation0.01
AI hardware0.00
Natural language0.00
Speech0.00
Planning0.00
Evolutionary computation0.00
Machine learning0.00

Ownership

TOYOTA JIDOSHA KABUSHIKI KAISHA

assignment · 223410486

Assignors

TSUKADA, AKIHIRO

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC