US5604823A — Method for extracting object images and method fo…

Claim 1

Independent Claim 248 words

A method for detecting a gradient of a contour line field, wherein a gradient of a contour line field is detected from information representing an object image falling within a region inside of a view window which has a predetermined size and is set in an image the view window having a size smaller than the entire image, the method for detecting the gradient of the contour line field comprises the steps of: i) cutting out a cut-out image which falls within the region inside of said view window having the predetermined size, from said image, said cut-out image corresponding to a candidate for a predetermined object image; ii) detecting from said cut-out image a contour line of said candidate for a predetermined object image, the contour line extending in a predetermined direction, the contour line being comprised of one or more components; iii) extracting from said detected contour line of said candidate for the predetermined object image all of said components of said detected contour line, which are tilted at a predetermined angle with respect to a direction along the circumference of a concentric circle centered on the center point of the view window; iv) detecting azimuths and intensities of said extracted components with respect to the center point of said view window, the azimuths and the intensities being detected as azimuth vectors; and v) composing a vector from said azimuth vectors, said vector corresponding to a gradient vector of the contour line field being thereby determined.

Claim 2

Independent Claim 248 words

A method for detecting a gradient of a contour line field, wherein a gradient of a contour line field is detected from information representing an object image falling within a region inside of a view window which has a predetermined size and is set in an image the view window having a size smaller than the entire image, the method for detecting the gradient of the contour line field comprises the steps of: i) cutting out a cut-out image which falls within the region inside of said view window having the predetermined size, from said image, said cut-out image corresponding to a candidate for a predetermined object image; ii) detecting from said cut-out image a contour line of said candidate for a predetermined object image, the contour line extending in a predetermined direction, the contour line being comprised of one or more components; iii) extracting from said detected contour line of said candidate for the predetermined object image all of said components of said detected contour line, which are tilted at a predetermined angle with respect to a direction along the circumference of a concentric circle centered on the center point of the view window; iv) detecting azimuths and intensities of said extracted components with respect to the center point of said view window, the azimuths and the intensities being detected as azimuth vectors; and v) composing a vector from said azimuth vectors, said vector corresponding to a gradient vector of the contour line field being thereby determined.

Claim 3

Dependent Claim 131 words

A method for detecting a gradient of contour line field as defined in claim 1, wherein the detection of said contour line of said candidate for the predetermined object image, which line extends in the predetermined direction, is performed after said cut-out image has been transformed with complex-log mapping into a complex-log mapped image by taking the center point of said view window as a pole of mapping; the extraction of said components of said detected contour line is performed by extracting from said detected contour line all contour line components which are tilted at a predetermined angle with respect to an annular direction in the complex-log mapped plane, and said azimuth vectors are detected by detecting azimuths and intensities of the extracted contour line components in said complex-log mapped plane.

Claim 4

Dependent Claim 42 words

A method for detecting a gradient of a contour line field as defined in claim 1, wherein a neural network is employed to perform at least a single step selected from the group consisting of steps i), ii), iii), iv) and v).

Claim 5

Dependent Claim 228 words

A method for detecting a gradient of a contour line field as defined in claim 1, wherein in step ii) a plurality of contour lines are detected from said cut-out image, said method further comprising the step of causing contour lines from among said plurality of detected contour lines, which have a high level of continuity in approximately the same direction as said predetermined direction and/or contour lines having a high intensity to cooperate with one another, thereby emphasizing said contour lines having a high level of continuity and high intensity; said contour lines having a high level of continuity and/or said contour lines having a high intensity and contour lines having a low level of continuity in approximately the same direction as said predetermined direction and/or contour lines having a low intensity are caused to compete with each other, whereby said contour lines having a low level of continuity and/or said contour lines having a low intensity are erased, and from an emphasized contour line, which has been formed by the emphasis of said contour lines having a high level of continuity and/or said contour lines having a high intensity, all components of said emphasized contour line which are tilted at a predetermined angle with respect to said direction along the circumference of a concentric circle centered on the center point of the view window, are extracted.

Claim 6

Dependent Claim 38 words

A method for detecting a gradient of a contour line field as defined in claim 4, wherein the emphasis and the erasing of said plurality of said detected contour lines are carried out by using a neural network.

Claim 7

Dependent Claim 31 words

A method for detecting a gradient of a contour line field as defined in claim 1, wherein in step v) phase shifts at predetermined angles are applied to said azimuth vectors.

Claim 8

Dependent Claim 46 words

A method for detecting a gradient of a contour line field as defined in claim 1, wherein in step v) weights are applied to said azimuth vectors in accordance with distances between the center point of the view window and said components of said contour line.

Claim 9

Dependent Claim 58 words

A method for detecting a gradient of a contour line field as defined in claim 1, wherein in step v) phase shifts at predetermined angles are applied to said azimuth vectors, and weights are applied to said azimuth vectors in accordance with distances between the center point of the view window and said components of said contour line.

Claim 10

Dependent Claim 43 words

A method for detecting a gradient of a contour line field as defined in claim 1, further comprising the step of moving the center point of the view window to a minimum point of the contour line field based on said gradient vector.

Claim 11

Dependent Claim 57 words

A method for detecting a gradient of a contour line field as defined in claim 9, further comprising the step of extracting a candidate for the predetermined object image in accordance with sizes and/or shapes of the candidate for the predetermined object image by taking said minimum point of the contour line field as a reference point.

Claim 12

Dependent Claim 33 words

A method for detecting a gradient of a contour line field as defined in claim 1, further comprising the step of: v) setting the size of the view window to another predetermined size.

Claim 13

Independent Claim 159 words

A method for detecting a gradient of an object image color field, wherein a gradient of the object image color field is detected based on information representing an object image falling within a region inside of a view window, the view window having a predetermined size and which is set within an image the view window having a size smaller than the entire image, the method for detecting the gradient of the object image color field comprising the steps of: i) cutting out from the image a cut-out image which falls within the region inside of said view window having the predetermined size; ii) extracting from said cut-out image a region which approximately coincides in color with said object image; iii) detecting an azimuth and a distance of said extracted region with respect to the center point of the view window; and iv) detecting said azimuth and said distance as a gradient vector of the object image color field.

Claim 14

Independent Claim 159 words

A method for detecting a gradient of an object image color field, wherein a gradient of the object image color field is detected based on information representing an object image falling within a region inside of a view window, the view window having a predetermined size and which is set within an image the view window having a size smaller than the entire image, the method for detecting the gradient of the object image color field comprising the steps of: i) cutting out from the image a cut-out image which falls within the region inside of said view window having the predetermined size; ii) extracting from said cut-out image a region which approximately coincides in color with said object image; iii) detecting an azimuth and a distance of said extracted region with respect to the center point of the view window; and iv) detecting said azimuth and said distance as a gradient vector of the object image color field.

Claim 15

Dependent Claim 316 words

A method for detecting a gradient of an object image color field as defined in claim 12, wherein regions from among a plurality of regions approximately coinciding in color with said object image, which exhibit a high degree of coincidence in color with said object image and are located at positions close to one another, are caused to cooperate with each other and are thereby emphasized, one of said regions which exhibits a high degree of coincidence in color with said object image, and one of said regions, which exhibits a low degree of coincidence in color with said object image and is located at a position spaced apart from said region exhibiting a high degree of coincidence in color with said object image, are caused to compete with each other, wherein said region which exhibits a low degree of coincidence in color with said object image is erased, said region which exhibits a high degree of coincidence in color with said object image and is located at a position spaced apart from another region which exhibits a high degree of coincidence in color of said object image, is caused to compete with the other region, wherein if said region exhibits a high degree of coincidence in color with said object image and has a size and a shape appropriate for the region to be selected, said region is kept unerased, and if said region exhibits a high degree of coincidence in color with said object image and has a size and a shape inappropriate for the region to be selected, said region is erased, whereby if said region is unerased and is a most appropriate region inside said view window, said region is selected as a predetermined object image region, and an azimuth and a distance of said selected predetermined object image region are detected with respect to the center point of said view window.

Claim 16

Dependent Claim 42 words

A method for detecting a gradient of an object image color field as defined in claim 12, wherein at least a single step selected from the group consisting of steps ii), iii) and iv) is carried out on a complex-log mapped plane.

Claim 17

Dependent Claim 48 words

A method for detecting a gradient of an object image color field as defined in claim 12, wherein during the step of detecting said gradient vector, weights are applied to said distance in accordance with a distance between the center point of said view window and said region.

Claim 18

Dependent Claim 44 words

A method for detecting a gradient of an object image color field as defined in claim 12, further comprising the step of moving the center point of the view window to a minimum point of the contour line field based on said gradient vector.

Claim 19

Dependent Claim 34 words

A method for detecting a gradient of an object image color field as defined in claim 12, further comprising the step of: v) setting the size of the view window to another predetermined size.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2000-02-18	2000-08-19	2001-02-18	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2004-02-18	2004-08-19	2005-02-18	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2008-02-18	2008-08-19	2009-02-18	$7700.00	Window Closed / Lapsed

Method for extracting object images and method for detecting movements thereof

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag