Method and system for automating an image rejection process

1. A method for automating an image rejection process, said method comprising:

extracting features from at least one image among a batch of images utilizing an LBP (Local Binary Pattern) operator to train a classifier, said features comprising texture, spatial structure, and image quality characteristics, wherein said LBP operator extracts several different orientations and types of edge features in said at least one image, giving equal priority for all patterns found;

calculating said features with respect to a test image for submission of said features to said classifier, given an operating point corresponding to a desired false positive rate;

generating at least one output from said classifier as a confidence value corresponding to a likelihood of at least one of the following: a license plate being absent in said image, said license plate being unreadable, or said license plate being obstructed; and

comparing said confidence value against a threshold to determine if said at least one image should be removed from a human review pipeline, thereby reducing images requiring human review.

2. The method of claim 1 wherein said classifier comprises a random forest classifier.

3. The method of claim 2 further comprising processing said at least one image using an LPR (License Plate Recognition) engine after capturing said at least one image via said image-capturing unit.

4. The method of claim 1 wherein extracting features from at least one image among a batch of images to train a classifier, said features comprising texture, spatial structure, and image quality characteristics, further comprises:

splitting said at least one image into an m×n grid of blocks; and

computing at least one LBP feature among said features for each block among said m×n grid of blocks to concatenate said at least one LBP feature into a single feature vector, said at least one LBP feature comprising said features including said texture, said spatial structure and said image quality characteristics, said at least one LBP feature comprising a local descriptor that assigns with respect to said texture an 8-bit texture value for each pixel in said at least one image, wherein said 8-bit texture value is based on a difference in values between a center pixel and eight neighbors of said center pixel at a specific radius.

5. The method of claim 1 further comprising initially capturing said at least one image via an image-capturing unit.

6. The method of claim 1 further comprising;

sweeping an image-based classifier across said at least one image to identify a center of highest confidence ROT with respect to said at least one image; and

computing a brightness, a contrast, and a character frequency for an ROI in said at least one image.

7. The method of claim 6 wherein said image-based classifier comprises a SNoW (Sparse Network of Winnows) classifier.

8. The method of claim 4 further comprising;

sweeping an image-based classifier across said at least one image to identify a center of highest confidence ROI with respect to said at least one image; and

computing a brightness, a contrast, and a character frequency for an ROI in said at least one image.

9. The method of claim 4 wherein said image-based classifier is swept in a 2D manner across said at least one image and a heat-map of a classifier response of said image-based classifier is then generated wherein said heat-ma is low-filtered to remove noise and then a highest point of the filtered heat-map is selected as said center of highest confidence ROI with respect to said at least one image.

10. A system for automating an image rejection process, said system comprising:

at least one processor; and

a memory comprising instructions stored therein, which when executed by said at least one processor, cause said at least one processor to perform operations comprising:

extracting features from at least one image among a batch of images utilizing an LBP (Local Binary Pattern) operator to train a classifier, said features comprising texture, spatial structure, and image quality characteristics, wherein said LBP operator extracts several different orientations and types of edge features in said at least one image, giving equal priority for all patterns found;

calculating said features with respect to a test image for submission of said features to said classifier, given an operating point corresponding to a desired false positive rate;

generating at least one output from said classifier as a confidence value corresponding to a likelihood of at least one of the following: a license plate being absent in said at least one image, said license plate being unreadable, or said license plate being obstructed; and

comparing said confidence value against a threshold to determine if said at least one image should be removed from a human review pipeline, thereby reducing images requiring human review.

11. The system of claim 10 wherein said classifier comprises a random forest classifier.

12. The system of claim 10 further comprising an image-capturing unit that communicates electronically with said at least one processor and said memory, wherein said operations further comprise initially capturing said at least one image via said image-capturing unit.

13. The system of claim 12 wherein said at least one processor processes said at least one image using an LPR engine after capturing said at least one image via said image-capturing unit.

14. The system of claim 10 further comprising:

an image-based classifier that sweeps across said at least one image to identify a center of highest confidence ROT with respect to said at least one image; and

wherein said operations further comprise computing a brightness, a contrast, and a character frequency for an ROT in said at least one image.

15. The system of claim 14 wherein said image-based classifier comprises a SNoW classifier.

16. The system of claim 14 wherein said image-based classifier comprises a SNoW classifier and wherein extracting said features comprising texture, spatial structure, and image quality characteristics from said at least one image, further comprises:

splitting said at least one image into an m×n grid of blocks; and

computing LBP features for blocks among said m×n grid of blocks to concatenate said LBP features.

17. The system of claim 10 wherein said instructions for extracting said features comprising texture, spatial structure, and image quality characteristics from said at least one image, further comprises instructions configured for:

splitting said at least one image into an m×n grid of blocks; and

computing at least one LBP feature among said features for each block among said m×n grid of blocks to concatenate said at least one LBP feature into a single feature vector, said at least one LBP feature comprising said features including said texture, said spatial structure and said image quality characteristics, said at least one LBP feature comprising a local descriptor that assigns with respect to said texture an 8-bit texture value for each pixel in said at least one image, wherein said 8-bit texture value is based on a difference in values between a center pixel and eight neighbors of said center pixel at a specific radius.

18. A system for automating an image rejection process, said system comprising:

at least one image-capturing unit;

at least one processor that communicates electronically with said at least one image-capturing unit; and

a memory comprising instructions stored therein, which when executed by said at least one processor, cause said at least one processor to perform operations comprising:

extracting features from at least one image among a batch of images utilizing an (LBP Local Binary Pattern) operator to train a classifier, said features comprising texture, spatial structure, and image quality characteristics, wherein said LBP operator extracts several different orientations and types of edge features in said at least one image, giving equal priority for all patterns found;

calculating said features with respect to a test image for submission of said features to said classifier, given an operating point corresponding to a desired false positive rate;

generating at least one output from said classifier as a confidence value corresponding to a likelihood of at least one of the following: a license plate being absent in said at least one image, said license plate being unreadable, or said license plate being obstructed; and

comparing said confidence value against a threshold to determine if said at least one image should be removed from a human review pipeline, thereby reducing images requiring human review.

19. The system of claim 18 wherein said classifier comprises a random forest classifier and wherein said at least one processor processes said at least one image using an LPR engine after capturing said at least one image via said image-capturing unit.

20. The system of claim 18 further comprising:

an image-based classifier that sweeps across said at least one image to identify a center of highest confidence ROT with respect to said at least one image; and

wherein said operations further comprise computing a brightness, a contrast, and a character frequency for an ROT in said at east one image.