Method of detecting a specific object in an image signal
AssigneeLG Electronics Inc.
InventorsJin Soo Lee, Hycon Jun Kim
Life of this patent
Estimating the term…
Abstract
A method of detecting a specific object in an image signal both efficiently and accurately is disclosed. According to the present invention, an object is first detected from an image or frame using a general feature of the object and in the following image or frame, the object is detected using an object-dependent feature. Also, when detecting a plurality of objects or portions of the object, a full color range of the specific object is determined, and color range sections of the full color range is used to detect the object.
Claims30
A method of detecting a specific object in an image signal comprising:
- detecting the specific object from an image using a general feature of the object;
- measuring a reliability of the detected object by a template matching;
- obtaining an object-dependent feature from the specific object detected in the detecting the specific object from an input image step;
- and
- detecting the specific object in the next input image using either the object-dependent feature or both the object-dependent feature and the general feature, wherein said template matching is performed by at least one of a whole matching method based on template matchings in predetermined positions, a motion estimation matching method based on a motion of an object, and a speed estimation matching method based on a relative speed difference of object motion appearing in the input images during a defined time interval (Δt).
A method of detecting a specific object in an image signal comprising:
- detecting the specific object from an input image using a general feature of the object;
- measuring a reliability of the detected object by a template matching;
- obtaining an object-dependent feature from the specific object detected in the detecting the specific object from an input image step;
- detecting the specific object in the next input image using either the object-dependent feature or both the object-dependent feature and the general feature, wherein said obtaining an object-dependent feature step is executed if the result of the template matching satisfies a predetermined threshold value.
A method, comprising:
- detecting a specific object from an input image using a general feature of the specific object;
- measuring an amount of reliability of the detected object by template matching;
- if the amount of reliability exceeds an predetermined value, then:
- obtaining an object dependent feature from the detected object;
- and
- detecting the specific object in another input image using the object-dependent feature either alone or in combination with the general feature, wherein the object-dependent feature is a subset of the general feature.
Dependencies are inferred from the claim text.