Imperium (IP) Holdings, Inc. v. Apple Inc. et al

Filing 1

COMPLAINT FOR PATENT INFRINGEMENT against Apple Inc., Kyocera Communications, Inc., LG Electronics Mobilecomm U.S.A., Inc., LG Electronics U.S.A., Inc., Motorola Mobility Holdings, Inc., Nokia Inc., Research In Motion Corporation, Sony Ericsson Mobile Communications (USA) Inc. ( Filing fee $ 350 receipt number 0540-2962750.), filed by Imperium (IP) Holdings, Inc.. (Attachments: # 1 Exhibit A - US Patent No. 6,271,884, # 2 Exhibit B - US Patent No. 6,838,651, # 3 Exhibit C - US Patent No. 6,838,715, # 4 Exhibit D - US Patent No. 7,064,768, # 5 Exhibit E - US Patent No. 7,109,535, # 6 Civil Cover Sheet)(Fisch, Alan)

Imperium (IP) Holdings, Inc. v. Apple Inc. et al

Doc. 1 Att. 4

Exhibit D

Dockets.Justia.com

(12)

United States Patent
Bao

(10) (45)

Patent NO.: US 7,064,768 B1 Date of Patent: Jun. 20,2006
A * 312000 Murphy ...................... 3821254 B l * 1212002 Granfors et al. ........... 378198.8 B1 * 512004 Horie et al. ................ 3821239 A1 * 912001 Gong et al. ................. 3451593

(54) BAD PIXEL CORRECTION WHILE PRESERVING FEATURES (75) Inventor: Yiliang Bao, Redondo Beach, CA (US) (73) Assignee: ESS Technology, Inc., Fremont, CA (US)
( * ) Notice: Subject to any disclaimer, the term of this patent is extended or adjusted under 35 U.S.C. 154(b) by 136 days.

6,038,031 6,498,831 6,735,341 200110020949

* cited by examiner
Primary Examiner-Kee M. Tung Assistant Examiner-Wesner Sajous (74) Attorney, Agent, or Firm-Farjami
& Farjami LLP

(57)

ABSTRACT

(21) Appl. No.: 101102,410 (22) Filed:
Mar. 20, 2002

(51) Int. C1. G09G 5/02 (2006.01) G06K 9/00 (2006.01) H04N 9/64 (2006.01) G09K 9/42 (2006.01) (52) U.S. C1. ...................... 3451589; 3451618; 3451698; 3821162; 3821254; 3481246 (58) Field of Classification Search ................ 3821149, 3821162-167,218, 168, 172, 254, 274, 276; 3481182, 189,272,246, 129, 134, 245, 247; 3451589, 593, 597, 612, 618, 698, 696 See application file for complete search history. (56)
5,432,863 A 5,764,209 A 6,035,072 A

References Cited
U.S. PATENT DOCUMENTS

* * *

711995 Benati et al. ............... 3821167 611998 Hawthorne et al. ........... 345187 312000 Read .......................... 3821275

A pixel correction system is provided. The pixel correction system includes a dynamic range detection system that receives test pixel data and adjacent pixel data and determines whether the test pixel data is within minimum pixel characteristic data and maximum pixel characteristic data of the adjacent pixel data. For example, if the pixel characteristic data is intensity, the dynamic range detection system detects bad pixels by identifying those pixels having an intensity value that is greater than the maximum intensity value of an adjacent pixel, or less than the minimum intensity value of an adjacent pixel. A dynamic range correction system coupled to the dynamic range detection system adjusts the test pixel data if the test pixel data is not within the maximum pixel characteristic data and the minimum pixel characteristic data, such as by setting the test pixel data equal to the maximum pixel characteristic data if the test pixel data is greater than the maximum pixel characteristic data, and by setting the test pixel data equal to the minimum pixel characteristic data if the test pixel data is less than the minimum pixel characteristic data.

20 Claims, 9 Drawing Sheets

Thresholds: 1. User-defined 2.a. Adaptively Calculated (Color Independent) 2.b. Adaptively Calculated (Color Dependent) 3. Adaptively Calculated (Greyscale) Saturation Re-Map Threshold

J
If

Bad Pixel Correction 300

Threshold

MAX 322
I
L

MAXs and MlNs are Function of Selected Neighboring Pixels,

, I
-IN= ReMap1
. .

Is
L

2-

2
Threshold ""8 ReMan 1

a

MAXs and MlNsare Pr~c----~

Threeod
Zero Intensity (Color Spectral Content or Other Image Data Characteristic)

Threshold

328

limb
5
3
(D

-----

0 L 1

-.

Re-Map,

.

310

Location 320

Pixel 3 Location 330

-

104 BAD PIXEL DETECTION SYSTEM MINIMUM CHARACTERISTIC SYSTEM 108 MAXIMUM CHARACTER1STIC SYSTEM COLOR ZONE SYSTEM

BAD PIXEL CORRECTION SYSTEM 106 CORRECTION

THRESHOLD

PIXEL TESTING

NON-COLOR ZONE SYSTEM

THRESHOLD CORRECTION

PIXEL ANALYSIS SYSTEM 102

PIXEL DATA INPUT

PIXEL DATA OUTPUT FIGURE 1

J

Bad Pixel Correction 200,

Bad Pixel Correction Multiple Correction Pixels Zone 1 201B

Multi-color Correction Pixel Zone 1 201A

Multi-color Correction Pixels Zone 2 202A

-

Multiple Correction Pixels - Zone 2 202B

Bayer Pattern

Arbitrary Pixel Pattern

FIGURE 2

FIGURE 3

J

Bad Pixel Correction 200,

J

Bad Pixel Correction 200,

Bayer Pattern

FIGURE 4

FIGURE 5

U.S. Patent

Jun. 20,2006

Sheet 4 of 9

J

Bad Pixel Correction 200,

r

8 Same Color Correction Pixels - Zone 1 201F
16 Same Color Correction Pixels - Zone 2 202F

f

8+16=24 Same Color Correction Pixels - Zone 3 (Use Both Zones 1 &2) 203F
---

Bayer Pattern
FIGURE 7

Thresholds:
1. User-defined 2.a. Adaptively Calculated (Color 1,ndependent) 2.b. Adaptively Calculated (Color Dependent) 3. Adaptively Calculated (Greyscale)

J

Bad Pixel Correction 300
T
b

Saturation Threshold Threshold 326

255

Re-Map

Threshold 336

MAX 322
I

MAX 332

+

-----

C

I I

MAXs and MlNs are Function of Selected Neighboring Pixels hreeiold 318

4

ty

MAXs and MlNs are Programmed
I

Zero lntensitv (Color spectrk
V", I.",
I,
-.I

Re- T / M~P/
Pixel I Location 310

Threshold 328 & R ~ - M ~ D ,k Pixel 2 Location 320
FIGURE 8

/I

ReMapi

'

Other Image Data Characteristic)

1 0 Pixel 3
Location 330

U.S. Patent

Jun. 20,2006

Sheet 7 of 9

-0

u

a,

t ?

m m
(\1

I-1

X

A

m u, m zF 4-0,

-~
I

0

I

I I

~

2 a

2 I 1
I

A

Unchanged

a

z

% 8 ~ : " 3

- '6 0 3

Unchanged
cn

" -so
C

U1

E
3

z

X o "-5 m

@((IN

(3
LL

$2

.x
.-

C

- -sz 0 0 0 m-7
0

,

t 0 .e

m

+

2 3
m

iG. n= s c a ,c $ 2 0 a +
O a,.?

a
2

z

a ,

0

U.S. Patent

Jun. 20,2006

Sheet 8 of 9

~p

-

Bad Pixel Correction Method
700

Select Zone/Trajectory/PixeIs 740 Calculate Correlation of Selected Pixels 750 Calculate Dymanic Range of Pixel as Function of Selected Zonerrrajectory Calculate Thresholds 760
I I

Detect Bad

No

- - - - - - - - - - - - -I
FIGURE 11

1
BAD PIXEL CORRECTION WHILE PRESERVING FEATURES
BACKGROUND OF THE INVENTION
5

2
pixel data equal to the maximum pixel characteristic data if the test pixel data is greater than the maximum pixel characteristic data, and by setting the test pixel data equal to the minimum pixel characteristic data if the test pixel data is less than the minimum pixel characteristic data. The present invention provides many important technical advantages. One important technical advantage of the present invention is a system and method for bad pixel detection and correction that use dynamic range thresholds to determine whether a pixel is bad. If the pixel is bad, then the pixel characteristic data is replaced with characteristic data for one of the adjacent pixels. The thresholds for the present invention can be adjusted to accommodate detail levels so as to provide bad pixel detection and correction that does not imnair detail resolution. Other systems, methods, features and advantages of the invention will be apparent to one with skill in the art upon examination of th; following figures and detailed desciiption. It is intended that all such additional systems, methods, features and advantages be included within this description, be within the scope of the invention, and be protected by the accompanying claims. BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention. In the figures, like reference numerals designate corresponding parts throughout the different views. FIG. 1is diagram of a system for detecting and correcting bad pixels in accordance with an exemplary embodiment of the present invention. FIG. 2 is a diagram of bad pixel detection and correction zones on a Bayer pattern in accordance with an exemplary embodiment of the present invention. FIG. 3 is a diagram of bad pixel detection and correction zones on an arbitrary pixel pattern in accordance with an exemplary embodiment of the present invention. FIG. 4 is a system diagram illustrating another exemplary embodiment of bad pixel detection and correction zones in accordance with the present invention. FIG. 5 is a system diagram illustrating another exemplary embodiment of bad pixel detection and correction zones. FIG. 6 is a diagram of a bad pixel detection and correction zone in accordance with an exemplary embodiment of the present invention. FIG. 7 is a diagram of a bad pixel detection and correction zone in accordance with another exemplary embodiment of the Present invention. FIG. 8 is a system diagram illustrating a bad pixel detection and correction process in accordance with an exemplary embodiment of the present invention. FIG. 9 is a system diagram illustrating a bad pixel detection and correction process in accordance with another exemplary embodiment of the present invention. FIG. 10 is a flowchart of a bad pixel detection and correction method in accordance with an exemplary embodiment of the present invention. FIG. 1 is a flowchart of a bad pixel detection and 1 correction method with trajectory selection in accordance with an exemplary embodiment of the present invention. A better understanding of the present invention can be obtained when the following detailed description of various exemplary embodiments is considered in conjunction with the following drawings.

1. Field of the Invention The invention relates to image processing and, more particularly, to bad pixel detection and correction within digital image processing that preserves features within the image. 2. Related Art Conventional bad pixel detection and correction systems often employ multi-stage median filtering. The multi-stage median filtering is performed as shown below. For the 3x3 pixel array: aO, a l , a2 a3, A, a4 a5, a6, a7 where aO, a l , a2, a3, a4, a5, a6, and a7 are the neighboring pixels, the multi-stage median filtering process is performed as follows: H=median (a3, A, a4) V=median (al, A, a6) O=median (H, A, V) SE=median (aO, A, a7) NE=median (a5, A, a2) D=median (SE, A, NE) At=median ( 0 , A, D) One of the deficiencies of this conventional method of bad pixel correction is that three operations are needed to determine each median, and a total of twenty-one operations are performed for each pixel. The multi-stage median filtering process also must be performed on every pixel, instead of just those that are actually bad. Another conventional method for correcting bad pixels is normal median filtering, but this method must also be performed on every pixel, and can undesirably smooth detail features on the image. SUMMARY OF THE INVENTION In accordance with the present invention, a system and method for bad pixel detection and correction are provided that overcome known problems with systems and methods for bad pixel detection and correction. In particular, a system and method for bad pixel detection and correction are provided that use dynamic range thresholds to determine whether a pixel is bad and to determine the correction factor. In accordance with an exemplary embodiment of the present invention, a pixel correction system is provided. The pixel correction system includes a dynamic range detection system that receives test pixel data and adjacent pixel data and determines whether the test pixel data is within minimum pixel characteristic data and maximum pixel characteristic data of the adjacent pixel data. For example, if the pixel characteristic data is intensity, the dynamic range detection system detects bad pixels by identifying those pixels having an intensity value that is greater than the maximum intensity value of an adjacent pixel plus a threshold value, or less than the minimum intensity value of an adjacent pixel minus a threshold value. A dynamic range correction system coupled to the dynamic range detection system adjusts the test pixel data if the test pixel data is not within the maximum pixel characteristic data and the minimum pixel characteristic data, such as by setting the test

lo

15

20

25

30

35

40

45

50

55

60

65

3
DETAILED DESCRIPTION OF THE INVENTION

4

can be created by data compression processes, noise introduced during data transmission, or in other manners. Dynamic range detection system 104 determines whether a FIG. 1 is diagram of a system 100 for detecting and test pixel is bad based on characteristic data from as few as correcting bad pixels in accordance with an exemplary 5 two adjacent pixels and generates bad pixel flag data. embodiment of the present invention. System 100 can use Dynamic range correction system 106 receives the bad the characteristic data, such as intensity or brightness, for as pixel flag data from dynamic range detection system 104 for few as two pixels adjacent to a test pixel to determine a test pixel and modifies the test pixel characteristic data to whether the test pixel is a bad pixel, and can also use the correct the bad pixel. In one exemplary embodiment, characteristic data for one of those adjacent pixels to deter- l o dynamic range correction system 106 can select pixel cormine the correction data for the pixel. rection data based on the pixel characteristic data from one adjacent pixel. System 100 includes pixel analysis system 102, which can Minimum characteristic system 108 and maximum charbe implemented in hardware, software, or a suitable combination of hardware and software, and which can be one or acteristic system 110 of dynamic range detection system 104 more hardware systems, or one or more software systems 1s determine the minimum and maximum pixel characteristic, operating on a suitable processing platform. As used herein, respectively, for a set of adjacent pixels. In one exemplary a hardware system can be one or more semiconductor embodiment, a set of adjacent pixels can be identified and devices, an application specific integrated circuit, a field minimum characteristic system 108 can determine the miniprogrammable gate array, or other suitable systems or commum characteristic value of the set, such as intensity. ponents. A software system can include one or more objects, 20 Likewise, maximum characteristic system 110 can deteragents, lines of code, threads, subroutines, databases, applimine the maximum characteristic value. The set of adjacent pixels can be selected by color zone system 112, non-color cation programming interfaces (APIs), or other suitable data structures, source code (human readable), object code (mazone system 118, or other suitable systems. In this exemchine readable), and can include two or more different lines plary embodiment, color zone system 112 can select a set of of code or suitable data structures operating in two or more 25 adjacent pixels based on the color of the test pixel, on the separate software applications, on two or more different pixel color distribution pattern type (such as a Bayer patprocessing platforms, or in other suitable architectures. In tern), or other suitable data. Likewise, non-color zone sysone exemplary embodiment, a software system can include tem can select a set of adjacent pixels based on the proximity of the adjacent pixels to the test pixel, regardless of color. one or more lines of code or other suitable software structures operating in a general purpose software application, 30 Other suitable processes can be used to select a set of such as an operating system, and one or more lines of code adjacent pixels from which to select the maximum and or other suitable software structures operating in a specific minimum characteristic value. purpose software application. In another exemplary embodiThreshold system 114 can add or subtract a threshold ment, a software system can be implemented as a distributed from the minimum and maximum characteristic value detersoftware system, on a different processing platform than that 35 mined by minimum characteristic system 108 and maximum shown in the exemplary embodiments herein, or in other characteristic system 110, respectively. In one exemplary suitable manners. embodiment, threshold system 114 can add the same quanPixel analysis system 102 is coupled to pixel data input tity to the maximum characteristic value as is subtracted and pixel data output. As used herein, the term "couple" and from the minimum characteristic value. In another exemits cognate terms such as "coupled" and "couples" can 40 plary embodiment, threshold system 114 can add a different include a physical connection (such as through a conducting quantity to the maximum characteristic value from what is layer of a semiconductor device), a virtual connection (such subtracted from the minimum characteristic value. In yet as through randomly assigned memory locations of a data another exemplary embodiment, threshold system 114 can memory device), a logical connection (such as through one add a quantity to the maximum characteristic value and or more logical devices of a semiconducting circuit), other 45 subtract a different quantity from the minimum characterissuitable connections, or a suitable combination of such tic value based on user-entered values, values from two or connections. In one exemplary embodiment, systems and more adjacent pixel zones, or other suitable criteria. components are coupled to other systems and components Pixel testing system 116 receives the test pixel data, the through intervening systems and components, such as minimum characteristic data, and the maximum characterthrough an operating system of a digital signal processor. 50 istic data, and determines whether the test pixel data is greater than the maximum characteristic data or less than the Pixel analysis system 102 includes dynamic range detection system 104, dynamic range correction system 106, minimum characteristic data. In one exemplary embodiment, pixel testing system 116 can use the maximum charminimum characteristic system 108, maximum characterisacteristic data and minimum characteristic data that has been tic system 110, color zone system 112, threshold system 114, pixel testing system 116, non-color zone system 118, mini 55 offset by the threshold data generated by threshold system 114. Pixel testing system 116 can then generate test result max correction system 120, and threshold correction system 122, each of which can be implemented in hardware, softdata that indicates whether the test pixel was within the ware, or a suitable combination of hardware and software, maximum characteristic data and the minimum characterisand which can be one or more hardware systems or one or tic data, was greater than the maximum characteristic data, more software systems operating on a suitable processing 60 or less than the minimum characteristic data. platform. Minimax correction system 120 and threshold correction Dynamic range detection system 104 receives pixel data system 122 receive the test result data and modify the test for a plurality of pixels and determines whether each pixel pixel data depending on the test result data. In one exemis "bad," meaning that it has characteristic data that is plary embodiment, minimax correction system 120 and defined in error. In one exemplary embodiment, a pixel can 65 threshold correction system 122 receive operating mode include an eight-bit characteristic, such as intensity, as well data, such that minimax correction system 120 is used to as a color spectral content or other suitable data. Bad pixels generate the modified test pixel data in a first mode and

US 7,064,768 B1
5

6

threshold correction system 122 is used to generate the 212D within an arbitrary pattern that are used to select pixels for performing bad pixel detection and correction for a modified test pixel data in a second mode. Minimax correction system 120 decreases the value of the test pixel data to center pixel R. the maximum characteristic data if the value of the test pixel FIG. 6 is a diagram of bad pixel correction zone 200E in data is greater than the maximum characteristic data. Like- 5 accordance with an exemplary embodiment of the present wise, minimax correction system 120 increases the value of invention. Bad pixel correction zone 201E demonstrates the test pixel data to the minimum characteristic data if the adjacent pixel selection for the 8 closest pixels when colorvalue of the test pixel data is less than the minimum dependent pixel characteristics are being tested for both red and blue pixels, although a red pixel is shown in bad pixel characteristic data. Threshold correction system 122 decreases the value of the test pixel data to the maximum l o correction zone 200E. Likewise, bad pixel correction zone 3 characteristic data or the maximum characteristic data plus 203E demonstrates adjacent pixel selection for up to the 24 the threshold data if the value of the test pixel data is greater closest pixels when color-dependent pixel characteristics are than the maximum characteristic data plus the threshold being tested for both red and blue pixels. FIG. 7 is a diagram of bad pixel correction zone 200F in data. Likewise, threshold correction system 122 increases the value of the test pixel data to the minimum characteristic 1s accordance with an exemplary embodiment of the present data or the minimum characteristic data minus the threshold invention. Bad pixel correction zone 201F demonstrates data if the value of the test pixel data is less than the adjacent pixel selection for the 8 closest pixels when colorminimum characteristic data minus the threshold data. dependent pixel characteristics are being tested for green pixels, which have a different density from that of red and operation, system 100 performs dynamic range bad pixel detection that detects whether a test pixel is a bad pixel 20 blue pixels in a Bayer pattern. Likewise, bad pixel ~ ~ r r e c t i o n based on pixel characteristic data of adjacent pixels, and zone 3 203F demonstrates adjacent pixel selection for up to dynamic range pixel correction that replaces the test pixel the 24 closest pixels when color-dependent pixel characterwith an appropriate adjacent pixel or other suitable data if istics are being tested for green pixels. the test pixel is determined to be a bad pixel. If the pixel FIG. 8 is a system diagram illustrating bad pixel COrreccharacteristic of the test pixel, such as intensity or color 25 tion Process 300 in xcordance with an exemplary embodiProcess spectral content, is greater than the largest characteristic ment of the Present invention. Bad pixel ~ ~ r r e c t i o n value of an adjacent pixel by more than a predetermined 300 determines a maximum value and a minimum value of amount, then the test pixel data is replaced with the data for a pixel characteristic for pixels adjacent to a test pixel, and of then determines thresholds based on the maximum value and that adjacent pixel, ~ i k ~if the ipixel ~ , ~ ~ the test pixel is less than the smallest characteristic value of 30 the minimum value to perform bad pixel detection and an adjacent pixel by more than a predetermined amount, correction. In another exemplary embodiment, the maxithen the test pixel data is replaced with the data for that mum value and the minimum value and the corresponding adjacent pixel. In this manner, pixels can be analyzed to thresholds can be predetermined or otherwise programmed. determine whether they are bad pixels, where the analysis is Bad pixel ~ ~ r r e c t i oProcess 300 includes exemplary n performed using a small number of operations and in a 35 pixel 1 location 310, pixel 2 location 320, and pixel 3 manner that does not result in a loss of detail. location 330. For pixel 1 location 310, maximum 312 with a 316 and minimum 314 with a FIG, 2 is a diagram of bad pixel detection and correction 318 are used to P ~ ~ ~ zones 200A on a Bayer pattern in accordance with an of the pixel 1 location 310 between a zero intensity level and exemplary embodiment of the present invention, Bad pixel detection and correction zones 200A include zone 201A 40 a saturation level in accordance with the present invention. In this the zero intensity level is encompassing the 8 pixels adjacent to the test pixel R, and shown as having a "0" value, and the saturation level is zone 202A encompassing the 16 pixels adjacent to the shown as having a "255" such as when an eight bit zone 1 pixels, In one exemplary embodiment, zone 1 201A image data system is used. The difference between the can be used to select adjacent pixels to analyze color312 and 316 can be to the independent pixel data characteristics, such as intensity, and 45 difference between threshold 318 and minimum 314. Likezone 202A can be used to select adjacent pixels to analyze wise, the differencebetween the maximum 312 and threshcolor-dependent characteristics, such as intensity based on old 316 can be different from the difference between threshcolor spectral content. old 318 and minimum 314, can be selected based on pixel FIG. 3 is a diagram of bad pixel detection and correction 50 variability characteristics for surrounding pixels, or can 200B On an pixel pattern in with be adjusted to accommodate for local image data an embodiment of the present invention. Bad variations, Threshold 316 and threshold 318 can be userpixel and 200B a test defined, adaptively calculated using different color adjacent pixel within an pixel pattern shown by a 5x5 pixels for color-independent pixel characteristics, adaptively array of pixels mmbered through 25. Zone 201B 55 calculated using same color adjacent pixels for color-depenencompasses the 8 pixels adjacent to the test pixel R, and dent pixel characteristics, or other suitable procedures can be zone 2 202B encompasses the 16 pixels adjacent to the zone Used to select or determine threshold 316 and 318, 1 pixels. When the pixel characteristic for pixel 1 location 310 is FIG. 4 is a system diagram illustrating another exemplary above the threshold 316 above the maximum 312, then the embodiment of bad pixel correction zones 200C in accor- 60 characteristic value of the pixel 1 location 310 is re-mapped dance with the present invention. Bad pixel correction zones to the maximum 312, another exemplary embodiment, the 200C include pixel trajectories 210C and 212C within a characteristic value of the pixel 1 location 310 can be Bayer Pattern that are used to select pixels for performing re-mapped to the threshold 316. Similarly, when the pixel bad pixel detection and correction for a center pixel R. characteristic is below the threshold 318 below the miniFIG. 5 is a system diagram illustrating another exemplary 65 mum 314, then the characteristic value of the pixel 1 embodiment of bad pixel correction zones 200D. Bad pixel location 310 is re-mapped to the minimum 314, or the correction zones 200D include pixel trajectories 210D and threshold 318 in another exemplary embodiment.

~

~

~

US 7,064,768 B1
7

8

Pixel 2 location 320 and pixel 3 location 330 demonstrate of detail that can be encountered using prior art methods by minimum, maximum, and corresponding threshold values assigning threshold values at a level that does not remove detail. for other exemplary pixels. Maximum 322 and minimum 324 of pixel 2 location 320 are different from maximum 312 FIG. 11 is a flowchart of a bad pixel correction method and minimum 314 of pixel 1 location, thus demonstrating 5 700 with trajectory selection in accordance with an exemhow the maximum and minimum values can be dynamically plary embodiment of the present invention. Method 700 assigned based on the values of adjacent pixels. Maximum begins at 710 where a dynamic range of a pixel characteristic 332 and minimum 334 of pixel 3 location 330 demonstrate for a test pixel is calculated as a function of pixel characa potential problem that can be encountered, wherein threshteristic data for pixels in a selected zone or trajectory. The old 336 of maximum 332 exceeds the saturation value for a l o method then proceeds to 720, where the test pixel is anapixel characteristic. In this exemplary embodiment, the lyzed to determine whether it is a bad pixel, such as by comparing the pixel characteristic to a maximum and minithreshold 336 is re-assigned to the saturation value. A similar process is implemented when the threshold 338 for the mum value to determine whether it is greater than the maximum or less than the minimum. The method then minimum 334 is less than the zero characteristic value. FIG. 9 is a system diagram illustrating bad pixel correc- 1s proceeds to 730 where it is determined whether the test pixel tion process 500 in accordance with an exemplary embodiis a bad pixel. If it is determined that the test pixel is not a ment of the present invention. Bad pixel correction process bad pixel, the method terminates for the current pixel, and 300 determines a maximum value and a minimum value of another pixel can be analyzed. a pixel characteristic for pixels adjacent to a test pixel, and If it is determined that the test pixel is a bad pixel, then but does not assign thresholds. Thus, bad pixel correction 20 the method proceeds to 735 where it is determined whether any thresholds used to perform bad pixel correction are process 500 can be used when a threshold is not used to detect and correct bad pixels. predetermined or not. If the thresholds are not predetermined the method proceeds to 740 where a selection is made of the FIG. 10 is a flowchart of a bad pixel correction method 600 in accordance with an exemplary embodiment of the zone, the trajectory, andor the pixels that are to be used to present invention. Method 600 begins at 610, where a 25 perform the necessary calculations to determine the threshdynamic range for a test pixel is detected. In one exemplary olds. The method then proceeds to 750 where the correlation embodiment, the range can be detected using the maximum of the selected pixels is calculated. The method then proand minimum values of a pixel characteristic for a set of ceeds to 760 where the thresholds are then calculated. The adjacent pixels. The method then proceeds to 620. method then proceeds to 765 where it is determined whether At 620, the test pixel is compared to the dynamic range. 30 the maximums and minimums are predetermined. In one exemplary embodiment, the characteristic for the test In one exemplary embodiment, the method proceeds from pixel can be compared to the maximum and minimum 760 to 790 where the new thresholds are re-mapped to the characteristic values for adjacent pixels, to maximum and old thresholds. In yet another exemplary embodiment, the minimum characteristic values for adjacent pixels that have pixels outside of the maximums and minimums are rebeen offset by a threshold amount, or other suitable values. 35 mapped to the maximums and minimums as shown in 795. The method then proceeds to 630, where it is determined If it is determined at 735 that the thresholds are predewhether the test pixel is a bad pixel. If it is determined that termined, then the method proceeds to 765 where it is the test pixel is not a bad pixel, then the method terminates determined whether the maximums and minimums are prefor that pixel, and the next test pixel can be processed. If it determined. If the maximums and minimums are predeteris determined at 630 that the test pixel is a bad pixel, then 40 mined, then the method proceeds to 780 and the pixels outside of the thresholds are re-mapped to the maximums the method proceeds to 635. and minimums. If the maximums and minimums are not At 635, it is determined whether different thresholds or minimumimaximum values are used to perform bad pixel predetermined, then the method proceeds to 770 where the correction as compared to those that are used to perform bad maximums and minimums are calculated. The pixels outside pixel detection. If the thresholds or minimumimaximum 45 of the thresholds are then re-mapped to the maximums and values are not different, the method either proceeds to 690, minimums as shown in 780, or the pixels outside of the where the pixels are remapped to the threshold values, or to maximums and minimums are mapped to the maximums 695, where the pixels are remapped to the corresponding and minimums as shown in 795. maximum or minimum value. In one exemplary embodiWhile various embodiments of the invention have been ment, if a pixel has a characteristic that is less than the 50 described, it will be apparent to those of ordinary skill in the minimum value for that characteristic or the minimum value art that many more embodiments and implementations are minus a threshold, it is mapped to the minimum value or the possible that are within the scope of this invention. minimum value minus the threshold. Likewise, if a pixel has What is claimed is: 1. A pixel correction system, comprising: a characteristic that is greater than the maximum value for that characteristic or the maximum value plus a threshold, it 55 a dynamic range detection system receiving test pixel data is mapped to the maximum value or the maximum value plus and adjacent pixel data and determining whether the the threshold. test pixel data is within minimum pixel characteristic data and maximum pixel characteristic data of the If it is determined at 635 that the thresholds are different, adjacent pixel data; and then the method proceeds to 640 where new thresholds or minimumimaximum values are calculated. The method then 60 a dynamic range correction system coupled to the dynamic range detection system, the dynamic range proceeds to 650 where the pixel is mapped to the new threshold or minimumimaximum value. correction system adjusting the test pixel data if the test pixel data is not within the maximum pixel characterIn operation, bad pixel correction method 600 provides for the detection and correction of bad pixels based on istic data and the minimum pixel characteristic data; wherein the dynamic range correction system further dynamic range thresholds. Method 600 thus allows bad pixel 65 correction to be performed using less processing resources comprises a minimax correction system adjusting the than other methods, and avoids the inadvertent elimination test pixel data to equal the minimum characteristic data

US 7,064,768 B1 9
10
if the test pixel data is less than the minimum characadjacent pixels comprises determining the maximum charteristic data and adjusting the test pixel data to equal the acteristic value of each of the adjacent pixels. 10. The method of claim 8 wherein determining the maximum characteristic data if the test pixel data is greater than the maximum characteristic data. maximum characteristic value of the characteristic data of 2. A pixel correction system, comprising: 5 the adjacent pixels comprises determining the maximum a dynamic range detection system receiving test pixel data characteristic value of each of the adjacent pixels and adding and adjacent pixel data and determining whether the a threshold value. 11. The method of claim 8 wherein determining the test pixel data is within minimum pixel characteristic data and maximum pixel characteristic data of the minimum characteristic value of the characteristic data of l o the adjacent pixels comprises determining the minimum adjacent pixel data; and a dynamic range correction system coupled to the characteristic value of each of the adjacent pixels. 12. The method of claim 8 wherein determining the dynamic range detection system, the dynamic range correction system adjusting the test pixel data if the test minimum characteristic value of the characteristic data of pixel data is not within the maximum pixel characterthe adjacent pixels comprises determining the minimum istic data and the minimum pixel characteristic data; 1s characteristic value of each of the adjacent pixels and wherein the dynamic range correction system further subtracting a threshold value. 13. A system for detecting a bad pixel comprising: comprises a threshold correction system adjusting the test pixel data to equal the minimum characteristic data a minimum characteristic system determining minimum minus a minimum threshold value if the test pixel data characteristic data for a plurality of adjacent pixels; a maximum characteristic system determining maximum is less than the minimum characteristic data and adjust- 20 characteristic data for the plurality of adjacent pixels; ing the test pixel data to equal the maximum characteristic data plus a maximum threshold value if the test and pixel data is greater than the maximum characteristic a pixel testing system receiving test pixel data, the minidata. mum characteristic data and the maximum character3. The system of claim 2, wherein the dynamic range 25 istic data and generating test result data as a function of whether the test pixel data is less than the minimum detection system further comprises a minimum characteristic system determining the minimum pixel characteristic characteristic data or greater than the maximum chardata from the adjacent pixel data. acteristic data, wherein the generating comprises 4. The system of claim 2, wherein the dynamic range adjusting the test pixel data to equal the minimum characteristic data if the test pixel data is less than the detection system further comprises a maximum character- 30 istic system determining the maximum pixel characteristic minimum characteristic data and adjusting the test pixel data from the adjacent pixel data. data to equal the maximum characteristic data if the test 5. The system of claim 2, wherein the dynamic range pixel data is greater than the maximum characteristic correction system further comprises a color zone system data. 14. The system of claim 13 further comprising a zone generating the adjacent pixel data based on test pixel color 35 data. system receiving the test pixel data and selecting the set of 6. The system of claim 2, wherein the dynamic range adjacent pixels. 15. A system for detecting a bad pixel comprising: correction system further comprises a non-color zone system generating the adjacent pixel data without regard to test a minimum characteristic system determining minimum 40 characteristic data for a plurality of adjacent pixels; pixel color data. 7. The system of claim 2, wherein the dynamic range a maximum characteristic system determining maximum detection system further comprises a pixel testing system characteristic data for the plurality of adjacent pixels; comparing the test pixel data to the minimum characteristic and data and the maximum characteristic data and generating a pixel testing system receiving test pixel data, the minitest result data that identifies whether the test pixel data is 45 mum characteristic data and the maximum characterless than the minimum pixel characteristic data or greater istic data and generating test result data as a function of than the maximum pixel characteristic data. whether the test pixel data is less than the minimum 8. A method for correcting pixels comprising: characteristic data or greater than the maximum charselecting two or more adjacent pixels to a test pixel, where acteristic data; and a green zone system receiving the test pixel data and each of the adjacent pixels and the test pixel have 50 associated characteristic data; selecting the set of adjacent pixels based on a green pixel distribution of a Bayer pattern. determining a maximum characteristic value of the char16. A system for detecting a bad pixel comprising: acteristic data of the adjacent pixels; a minimum characteristic system determining minimum determining a minimum characteristic value of the characteristic data of the adjacent pixels; 55 characteristic data for a plurality of adjacent pixels; correcting the test pixel characteristic data if it is less than a maximum characteristic system determining maximum the minimum characteristic data or greater than the characteristic data for the plurality of adjacent pixels; maximum characteristic data, wherein said correcting and comprises adjusting the test pixel characteristic data to a pixel testing system receiving test pixel data, the miniequal the minimum characteristic value if the test pixel 60 mum characteristic data and the maximum charactercharacteristic data is less than the minimum characteristic data and generating test result data as a function of istic data and adjusting the test pixel characteristic data whether the test pixel data is less than the minimum to equal the maximum characteristic data if the test characteristic data or greater than the maximum charpixel characteristic data is greater than the maximum acteristic data; and 65 a red zone system receiving the test pixel data and characteristic data. 9. The method of claim 8 wherein determining the maxiselecting the set of adjacent pixels based on a red pixel mum characteristic value of the characteristic data of the distribution of a Bayer pattern.

US 7,064,768 B1
11
17. A method, comprising: selecting two or more adjacent pixels to a test pixel, where each of the adjacent pixels and the test pixel have associated characteristic data; determining a maximum characteristic value of the characteristic data of the adjacent pixels; determining a minimum characteristic value of the characteristic data of the adjacent pixels; correcting the test pixel characteristic data if it is less than the minimum characteristic data or greater than the maximum characteristic data, wherein the correcting comprises adjusting the test pixel characteristic data to equal the minimum characteristic data minus a minimum threshold value if the test pixel characteristic data is less than the minimum characteristic data and adjusting the text pixel characteristic data to equal the maximum characteristic data plus a maximum threshold value if the test pixel characteristic data is greater than the maximum characteristic data. 18. A system, comprising: means for selecting two or more adjacent pixels to a test pixel, where each of the adjacent pixels and the test pixel have associated characteristic data; means for determining a maximum characteristic value of the characteristic data of the adjacent pixels; means for determining a minimum characteristic value of the characteristic data of the adjacent pixels; means for correcting the test pixel characteristic data if it is less than the minimum characteristic data or greater than the maximum characteristic data, wherein the correcting comprises adjusting the test pixel characteristic data to equal the minimum characteristic data minus a minimum threshold value if the test pixel characteristic data is less than the minimum characteristic data and adjusting the text pixel characteristic data to equal the maximum characteristic data plus a maximum threshold value if the test pixel characteristic data is greater than the maximum characteristic data.

12
19. A system, comprising: means for selecting two or more adjacent pixels to a test pixel, where each of the adjacent pixels and the test pixel have associated characteristic data; means for determining a maximum characteristic value of the characteristic data of the adjacent pixels; means for determining a minimum characteristic value of the characteristic data of the adjacent pixels; means for correcting the test pixel characteristic data if it is less than the minimum characteristic data or greater than the maximum characteristic data, wherein said correcting comprises adjusting the test pixel characteristic data to equal the minimum characteristic value if the test pixel characteristic data is less than the minimum characteristic data and adjusting the test pixel characteristic data to equal the maximum characteristic data if the test pixel characteristic data is greater than the maximum characteristic data. 20. A system, comprising: a minimum characteristic system determining minimum characteristic data for a plurality of adjacent pixels; a maximum characteristic system determining maximum characteristic data for the plurality of adjacent pixels; and a pixel testing system receiving test pixel data, the minimum characteristic data and the maximum characteristic data and generating test result data as a function of whether the test pixel data is less than the minimum characteristic data or greater than the maximum characteristic data, wherein the generating comprises adjusting the test pixel data to equal the minimum characteristic data minus a minimum threshold value if the test pixel data is less than the minimum characteristic data and adjusting the text pixel data to equal the maximum characteristic data plus a maximum threshold value if the test pixel data is greater than the maximum characteristic data.

5

10

15

20

25

30

35

* * * * *