US5263098A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Object recognition method and apparatus for determining the presence and shape of an object
Assignee: Panasonic Holdings Corp
•
Filing: 1991-07-08
•
Granted: 1993-11-16
•
Expires: 2011-07-08
Abstract
Abstract An object recognition apparatus for determining the presence and shape of an object in an image has a multiplier, an adder and a memory for obtaining histograms relative to densities for each of specified partial areas in the image. The histogram of each partial area is stored in the memory at addressed determined by adding an actual density of each of pixels in the partial area to a number indicating the partial area multiplied to be shifted to the left by the number of quantization bits. In a micro computer of the apparatus, the histogram for each partial area compared with a corresponding reference histogram which indicates that an object is present in an at least one part of the partial area. Based on a comparison result, the micro computer determines whether the object is present or not in each partial area. Then, based on the combination of partial areas where it is determined that the object is present, the shape of the object is evaluated.
Inventors
- 👤 Kinji Horikami
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-05 12:28
CPC Classifications
G; G06; G06V; G06V10/00; G06V10/40; G06V10/50; G06V10/507; G06T; G06T5/00; G06T5/40; G06T7/00; G06T7/10; G06T7/11
Claims Analysis
Total Claims: 12
Claim 1
Independent Claim
217 words
An object recognition apparatus which processes pixels positioned in a plurality of predetermined partial areas of an image containing an object to be recognized for determination whether the object is present or absent in each of the partial areas, comprising: first means for preparing a histogram of densities of whole pixels for each partial area while the image is being scanned, wherein the first means includes a histogram memory which has addresses defined by the combinations of numbers indicating the partial areas and all possible densities of the pixels in the partial areas, the histogram memory being initialized before the preparation of the histogram is started so that data at each address represents a value of zero, and when, in each partial area, an actual density is obtained, data at the address determined by the combination of the number representing the partial area and the actual density is read out from the histogram memory and incremented by one and the so incremented data is stored in the histogram memory at the same address; second means for comparing the histogram obtained by the first means for each partial area with reference data; and third means for determining whether the object is present or absent in each partial area based on a comparison result obtained by the second means.
Claim 2
Dependent Claim
217 words
An object recognition apparatus which processes pixels positioned in a plurality of predetermined partial areas of an image containing an object to be recognized for determination whether the object is present or absent in each of the partial areas, comprising: first means for preparing a histogram of densities of whole pixels for each partial area while the image is being scanned, wherein the first means includes a histogram memory which has addresses defined by the combinations of numbers indicating the partial areas and all possible densities of the pixels in the partial areas, the histogram memory being initialized before the preparation of the histogram is started so that data at each address represents a value of zero, and when, in each partial area, an actual density is obtained, data at the address determined by the combination of the number representing the partial area and the actual density is read out from the histogram memory and incremented by one and the so incremented data is stored in the histogram memory at the same address; second means for comparing the histogram obtained by the first means for each partial area with reference data; and third means for determining whether the object is present or absent in each partial area based on a comparison result obtained by the second means.
Claim 3
Dependent Claim
38 words
The object recognition apparatus as claimed in claim 1, further comprising means for evaluating the shape of the object from the combination of partial areas where it is determined by the third means that the object is present.
Claim 4
Dependent Claim
36 words
The object recognition apparatus as claimed in claim 1, wherein the reference data for each partial area is a reference histogram which indicates that the object occupies at least one part of the corresponding partial area.
Claim 5
Dependent Claim
36 words
The object recognition apparatus as claimed in claim 2, wherein the reference data for each partial area is a reference histogram which indicates that the object occupies at least one part of the corresponding partial area.
Claim 6
Dependent Claim
29 words
The object recognition apparatus as claimed in claim 3, further comprising means for normalizing the histogram for each partial area and the reference histogram relative to frequency and density.
Claim 7
Dependent Claim
30 words
The object recognition apparatus as claimed in claim 4, further comprising means for normalizing the histogram for each partial area and the corresponding reference histogram relative to frequency and density.
Claim 8
Dependent Claim
78 words
The object recognition apparatus as claimed in claim 5, further comprising means for accumulating differences in frequency between a normalized histogram for each partial area and the corresponding reference histogram at all densities, wherein the second means compares accumulated differences for each partial area with the corresponding reference value and the third means determines that the object is present in the corresponding partial area when a value representing the accumulated differences is smaller than a further reference value.
Claim 9
Dependent Claim
78 words
The object recognition apparatus as claimed in claim 6, further comprising means for accumulating differences in frequency between a normalized histogram for each partial area and the corresponding reference histogram at all densities, wherein the second means compares accumulated differences for each partial area with the corresponding reference value and the third means determines that the object is present in the corresponding partial area when a value representing the accumulated differences is smaller than a further reference value.
Claim 10
Dependent Claim
215 words
An object recognition method for processing pixels positioned in a plurality of predetermined partial areas of an image containing an object to be recognized for determination whether the object is present or absent in each of the partial areas, comprising steps of: preparing a histogram of densities of whole pixels for each partial area while the image is being scanned, wherein a histogram memory which has addresses defined by the combinations of numbers indicating the partial areas and all possible densities of the pixels in the partial areas is used, the histogram memory being initialized before the preparation of the histogram is started so that data at each address represents a value of zero, and when, in each partial area, an actual density is obtained, data at the address determined by the combination of the number representing the partial area and the actual density is read out from the histogram memory to be incremented by one and the so incremented data is stored in the histogram memory at the same address; comparing the histogram for each partial area with a reference histogram which indicates that the object occupies at least one part of the corresponding partial area; and determining whether the object is present or absent in each partial area based on a comparison result.
Claim 11
Dependent Claim
215 words
An object recognition method for processing pixels positioned in a plurality of predetermined partial areas of an image containing an object to be recognized for determination whether the object is present or absent in each of the partial areas, comprising steps of: preparing a histogram of densities of whole pixels for each partial area while the image is being scanned, wherein a histogram memory which has addresses defined by the combinations of numbers indicating the partial areas and all possible densities of the pixels in the partial areas is used, the histogram memory being initialized before the preparation of the histogram is started so that data at each address represents a value of zero, and when, in each partial area, an actual density is obtained, data at the address determined by the combination of the number representing the partial area and the actual density is read out from the histogram memory to be incremented by one and the so incremented data is stored in the histogram memory at the same address; comparing the histogram for each partial area with a reference histogram which indicates that the object occupies at least one part of the corresponding partial area; and determining whether the object is present or absent in each partial area based on a comparison result.
Claim 12
Dependent Claim
35 words
The object recognition method as claimed in claim 9, further comprising a step for evaluating the shape of the object from the combination of partial areas where it is determined that the object is present.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1996-11-15 | 1997-05-17 | 1997-11-16 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2000-11-15 | 2001-05-17 | 2001-11-16 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2004-11-15 | 2005-05-17 | 2005-11-16 | $7700.00 | Window Closed / Lapsed |