Patent US5453167A successfully refreshed from Google Patents.
Patent US5559899A successfully refreshed from Google Patents.
US5559899A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method for the adaptive quantization of a range of input values
Assignee: Robert Bosch GmbH
•
Filing: 1993-12-09
•
Granted: 1996-09-24
•
Expires: 2013-12-09
Abstract
Abstract A method for the adaptive quantization of the range of input values on the basis of the histogram method for estimating density. Input values are read-in and assigned to a quantization of the range of input values. From the distribution of the input values, a density-estimate function is determined which is transformed into a monotonically rising mapping function. The range of values of the mapping function is divided into equally sized sub-ranges which are translated into corresponding partitions in the range of input values. A new quantization of the range of input values is achieved which distributes the limited number of partitions so that a large number of small partitions are assigned to ranges of large fluctuations in the input-value distribution, and a small number of large partitions are assigned to ranges of small fluctuations in the input-value distribution.
Inventors
- 👤 Hauke Schmidt
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 14:49
CPC Classifications
G; G06; G06F; G06F17/00; G06F17/10; G06F17/18; G06F18/00; G06F18/20; G06F18/23; G06F18/232; G06F18/2321
Claims Analysis
Total Claims: 18
Claim 1
Independent Claim
121 words
A method for classifying objects, comprising the steps of: measuring a parameter of each of a plurality of objects; dividing a range of parameter values into partitions; assigning measured parameter values to the partitions; determining a density estimate for each partition based upon a relative frequency with which the measured parameter values are assigned to the partitions; based on the density estimates, redividing the range of parameter values into new partitions, each new partition having a plurality of new density estimates, such that a difference between two of the new density estimates of a new partition does not exceed a predetermined value; and classifying each object in accordance with the new partition in which the measured parameter of the object falls.
Claim 2
Independent Claim
121 words
A method for classifying objects, comprising the steps of: measuring a parameter of each of a plurality of objects; dividing a range of parameter values into partitions; assigning measured parameter values to the partitions; determining a density estimate for each partition based upon a relative frequency with which the measured parameter values are assigned to the partitions; based on the density estimates, redividing the range of parameter values into new partitions, each new partition having a plurality of new density estimates, such that a difference between two of the new density estimates of a new partition does not exceed a predetermined value; and classifying each object in accordance with the new partition in which the measured parameter of the object falls.
Claim 3
Dependent Claim
79 words
The method according to claim 1, further comprising the steps of: storing a count value for each partition in a storage device; incrementing by one the count values of the partitions to which a measured parameter value was assigned; and determining the density estimates of the partitions by dividing the stored count values by the total number of measured parameter values and by the size of the corresponding partitions; and replacing the density estimates with the new density estimates.
Claim 4
Dependent Claim
97 words
The method according to claim 1, further comprising the steps of: low-pass filtering the density estimates into a continuous estimated-value function; performing a subtraction operation on the continuous estimated-value function; generating an absolute value function from the subtracted continuous estimated-value function; performing a summation operation on the absolute value function to generate a mapping function; normalizing the range of the mapping function; dividing the range of the mapping function into a predetermined number of sub-ranges; and translating the sub-ranges into new partitions in the range of measured parameter values, using an inverse function of the mapping function.
Claim 5
Dependent Claim
41 words
The method according to claim 2, further comprising the step of: weighting the stored count values by a factor with a value ≧0 and ≦1 based upon overlapping of the partitions with the new partitions and a change in the partitions.
Claim 6
Dependent Claim
55 words
The method according to claim 3, wherein the dividing step includes the sub-step of dividing the range of function values of the mapping function into a fixed number of sub-ranges with a bottom and top sub-range and at least two equally sized middle sub-ranges, the bottom and top sub-ranges being smaller than the middle sub-ranges.
Claim 7
Dependent Claim
19 words
The method according to claim 3, wherein the low-pass filtering is carried out only at midpoints of the partitions.
Claim 8
Dependent Claim
49 words
The method according to claim 3, wherein the low-pass filtering is performed at a cut-off frequency which is determined by calculating a variance between the density estimates and the filtered density estimates and setting the cut-off frequency so that a maximum variance does not exceed a predetermined variance value.
Claim 9
Dependent Claim
18 words
The method according to claim 7, wherein the variance is determined only at the midpoints of the partitions.
Claim 10
Dependent Claim
13 words
The method according to claim 3, wherein the mapping function is low-pass filtered.
Claim 11
Dependent Claim
35 words
The method according to claim 3, wherein new partition boundaries are determined by weighting changes between new and old partitions by a factor ≧0 and ≦1, and adding the weighted changes to old partition boundaries.
Claim 12
Dependent Claim
18 words
The method according to claim 3, wherein the subtraction operation includes differentiation, and the summation operation includes integration.
Claim 13
Dependent Claim
21 words
The method according to claim 3, wherein the absolute-value generating step includes the sub-steps of squaring and generating a square root.
Claim 14
Dependent Claim
20 words
The method according to claim 5, wherein the bottom and top sub-ranges are one-half the size of the middle sub-ranges.
Claim 15
Independent Claim
125 words
A system for adaptively quantizing a range of input values, comprising: an input interface for reading-in input values; a processor for: dividing a range of input values into partitions; assigning input values to the partitions; determining a density estimate for each partition based upon a relative frequency with which the input values are assigned to the partitions; redividing the range of input values into new partitions based on the density estimates, each new partition having a plurality of new density estimates, such that a difference between two of the new density estimates of a new partition does not exceed a predetermined value; classifying at least one object as a function of the new partition; and an output interface for outputing the new density estimate values.
Claim 16
Independent Claim
125 words
A system for adaptively quantizing a range of input values, comprising: an input interface for reading-in input values; a processor for: dividing a range of input values into partitions; assigning input values to the partitions; determining a density estimate for each partition based upon a relative frequency with which the input values are assigned to the partitions; redividing the range of input values into new partitions based on the density estimates, each new partition having a plurality of new density estimates, such that a difference between two of the new density estimates of a new partition does not exceed a predetermined value; classifying at least one object as a function of the new partition; and an output interface for outputing the new density estimate values.
Claim 17
Independent Claim
127 words
A system for adaptively quantizing a range of input values, comprising: input means for reading-in input values; dividing means for dividing a range of input values into partitions; assigning means for assigning input values to the partitions; determining means for determining a density estimate for each partition based upon a relative frequency with which the input values are assigned to the partitions; redividing means for redividing the range of input values into new partitions, based on the density estimates, with each new partition having a plurality of new density estimates such that a difference between two of the new density estimates of a new partition does not exceed a predetermined value; and classifying means for classifying at least one object as a function of the new partition.
Claim 18
Independent Claim
127 words
A system for adaptively quantizing a range of input values, comprising: input means for reading-in input values; dividing means for dividing a range of input values into partitions; assigning means for assigning input values to the partitions; determining means for determining a density estimate for each partition based upon a relative frequency with which the input values are assigned to the partitions; redividing means for redividing the range of input values into new partitions, based on the density estimates, with each new partition having a plurality of new density estimates such that a difference between two of the new density estimates of a new partition does not exceed a predetermined value; and classifying means for classifying at least one object as a function of the new partition.
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) | 1999-09-24 | 2000-03-25 | 2000-09-24 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-09-24 | 2004-03-25 | 2004-09-24 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-09-24 | 2008-03-25 | 2008-09-24 | $7700.00 | Window Closed / Lapsed |