Image processing using histograms

• US 10,706,516 B2
• Filed: 09/15/2017
• Issued: 07/07/2020
• Est. Priority Date: 09/16/2016
• Status: Active Grant

• Current Assignee: Teledyne FLIR LLC (Teledyne Technologies Incorporated)
• Original Assignee: FLIR Systems, Inc. (Teledyne Technologies Incorporated)
• Inventors: Schulte, Stefan
• Primary Examiner(s): Chawan, Sheela C
• Application Number:
  US15/706,369
• Publication Number:
  US 20180082410A1
• Time in Patent Office: 1,026 Days
• Field of Search: 382170, 382275, 382131, 382293, 382274, 382100, 382276, 382254, 382128, 382168, 348 36, 348143, 34820799, 348335, 600173, 600425, 600111
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 7/75 involving models
• CPC Class Codes: G06V 10/32 Normalisation of the pattern dimensions
• CPC Class Codes: G06V 10/507 Summing image-intensity values; Histogram projection analysis
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 10/761 Proximity, similarity or dissimilarity measures

Abstract

A system and method for comparing and searching for digital data, such as images, using histograms includes receiving a source image, receiving a comparison image, generating a first histogram for the source image and generating a second histogram for the comparison image. The source image may be received from a network device, such as a computer or camera, and the comparison image may one of a plurality of stored images in a database. The histograms may correspond to an image characteristic, including a color histogram corresponding to the distribution of the intensity of a corresponding color among image pixels in the source image. Each of first and second histograms is normalized and a similarity score is calculated between the two histograms. The similarity score represents a similarity measure between the two histograms, calculated from a subset of bins, which are independently selected for each image.

Claims

1. A method for comparing two images comprising:

receiving a source image;

receiving a comparison image;

generating a first histogram for the source image;

generating a second histogram for the comparison image;

separately normalizing each of the first and second histograms to produce a normalized first histogram and a normalized second histogram; and

calculating a similarity score between the first normalized histogram and the second normalized histogram, the similarity score representative of a similarity measure between the first normalized histogram and the second normalized histogram.

2. The method of claim 1, wherein generating a first histogram for the source image comprises:

generating a color histogram corresponding to the color distribution of the image pixels.

3. The method of claim 2, wherein generating a first histogram for the source image further comprises:

generating a color histogram for each color component of the image, each color histogram representing the distribution of the intensity of the corresponding color among image pixels.

4. The method of claim 3, wherein each of the color histograms is normalized.

10. The method of claim 1, wherein the source image is received from a video camera and the comparison image is a stored background image;

and

wherein the similarity score is compared to a threshold value to detect whether a foreground object is present in the source image.

11. The method of claim 1, wherein the source image is a detected foreground object and the comparison image represents a known object from an object classification system;

and

wherein the similarity score is further representative of a similarity between the detected foreground object and the known object.

12. An image retrieval system configured to perform the method of claim 1, the image retrieval system comprising:

an image server comprising image storage and indexing logic and image retrieval logic; and

an image storage system storing image data including color histograms corresponding to stored images.

5. The method of 1, wherein calculating the similarity score further comprises calculating a similarity score from a subset of bins.

6. The method of claim 5, wherein the subset of bins is selected from the bins having the largest values.

7. The method of claim 6, wherein a predetermined number of the largest bins are selected.

8. The method of claim 6, wherein the largest bins are selected until a cumulative threshold is exceeded.

9. The method of claim 5 wherein calculating the similarity score further comprises calculating a total value of the subset of bins from each histogram and performing an overlap calculation representing matching bins pairs found in the selected subsets of bins.

13. A system comprising:

an image storage system configured to store image data including stored images and corresponding histograms;

a program memory storing program instructions; and

a processor configured to execute the program instructions;

to conduct a query for an image stored in the image storage system, wherein the processor is further configured to;

receive a source image;

determine a histogram for the source image representing a distribution of image characteristics;

normalize the histogram; and

calculate a similarity measure by comparing the normalized histogram of the source image with a normalized histogram corresponding to the one or more stored images.

14. The system of claim 13, wherein the

program memory is further configured to store instructions for storing, indexing, and retrieving images;

and

wherein the processor is further configured to;

receive an image for storing in the image storage system;

store the image in the image storage system;

calculate a histogram representing a distribution of image characteristics;

normalize the histogram;

store the normalized histogram in the image storage system and associated with the corresponding image.

15. The system of claim 13, wherein the processor is further configured to:

select a subset of bins from each histogram, the subset of bins representing the largest bin values in each histogram;

wherein the similarity measure is determined by comparing the subset of bins.

16. The system of claim 15, wherein the processor is further configured to:

select the n largest bins, where n is a predetermine number of bins.

17. The system of claim 15, wherein the processor is further configured to select the largest bin until the cumulative value of the selected bins exceeds a threshold value.

18. The system of claim 15, wherein the similarity measure includes a total value of the selected bins.

19. The system of claim 15, wherein the similarity measure includes an overlap calculation representing matching bins pairs found in the selected subsets of bins.

20. The system of claim 15, wherein the similarity measure is equivalent to 2 times the sum of the min value of each pair divided by the total value of the subsets.