System and method for identifying defects in welds by processing X-ray images

1. A method comprising:

capturing an X-Ray image of a weld joint, wherein the X-Ray image comprises dark areas and bright areas, and wherein the dark areas and the bright areas have gray scale information that indicates a luminance intensity;

executing a gamma correction algorithm on the X-Ray image by using a standard power law transform to enhance a contrast of the dark areas and the bright areas, wherein the gamma correction algorithm facilitates a differentiation between the dark areas and bright areas, and wherein the gamma correction algorithm is executed on the gray scale information associated with each pixel in the X-Ray image [p min , p max ], and wherein p min indicates the gray scale information of a darkest pixel in the X-Ray image, and wherein p max indicates the gray scale information of a brightest pixel in the X-Ray image;

passing, upon execution of the gamma correction algorithm, the X-Ray image through one or more de-noising filters in order to remove noise from the X-Ray image;

identifying, upon removing the noise, a fragment of the X-Ray image that comprises at least one of the dark area with respect to one or more neighboring areas, wherein the at least one of the dark areas is identified by using a gradient discontinuity algorithm, and wherein the at least one of the dark areas indicates a presence of a defect in the weld joint; and

segmenting the X-Ray image into one or more segments based on the identification of the at least one of the dark areas in the X-Ray image, wherein the segmenting facilitates an identification of the defect in the weld joint, and wherein the capturing, the executing, the passing, the identifying, the segmenting, or a combination thereof are performed by a processor.

2. The method of claim 1 , wherein the executing further comprises:

computing gamma correction values in the X-Ray image by using a mathematical formulation, wherein the mathematical formulation is log(255)/log(p max −p min ); and

transforming the gray scale information of each pixel of the X-Ray image using the standard power law thereby enhancing the contrast of the dark areas and the bright areas in the X-Ray image.

3. The method of claim 1 , wherein the X-Ray is segmented by using at least one of a region-based segmentation and a thresholding technique.

4. The method of claim 1 , wherein the one or more de-noising filters comprises a mean based de-noising, a weighted mean based de-noising, and an adaptive median based de-noising.

5. The method of claim 1 , wherein the identifying further comprises:

comparing the gray scale information of the at least one of the dark areas with respect to the one or more neighboring areas to identify the defect in the weld joint.

6. The method of claim 1 further comprising classifying the weld joint in at least one category of a defective category and a non-defective-category.

7. The method of claim 6 , wherein the weld joint is classified based upon the identification of the at least one of the dark areas present in the one or more segments of the X-Ray image.

8. A system comprising:

a processor; and

a memory coupled to the processor, wherein the processor is capable of executing a plurality of modules stored in the memory, and wherein the plurality of modules comprises:

an image capturing module configured to capture an X-Ray image of a weld joint, wherein the X-Ray image comprises dark areas and bright areas, and wherein the dark areas and the bright areas have gray scale information that indicates a luminance intensity;

an image processing module configured to

execute a gamma correction algorithm on the X-Ray image by using a standard power law transform to enhance a contrast of the dark areas and the bright areas, wherein the gamma correction algorithm facilitates a differentiation between the dark areas and bright areas, wherein the gamma correction algorithm is executed on the gray scale information associated with each pixel in the X-Ray image [p min , p max ], wherein p min indicates the gray scale information of a darkest pixel in the X-Ray image, and wherein p max indicates the gray scale information of a brightest pixel in the X-Ray image; and

pass, upon execution of the gamma correction algorithm, the X-Ray image through one or more de-noising filters in order to remove noise from the X-Ray image; and

an analysis module configured to

identify, upon removing the noise, a fragment of the X-Ray image that comprises at least one of the dark areas with respect to one or more neighboring areas, wherein the at least one of the dark areas is identified by using a gradient discontinuity algorithm, and wherein the at least one of the dark areas indicates a presence of a defect in the weld joint; and

segment the X-Ray image into one or more segments based on the identification of the at least one of the dark areas in the X-Ray image, wherein the segmenting facilitates an identification of the defect in the weld joint.

9. The system of claim 8 , further comprising a classification module configured to classify the X-Ray image of the weld joint in at least one category of a defective category and a non-defective-category.

10. The system of claim 9 , wherein the weld joint is classified based upon the identification of the at least one of the dark areas present in the one or more segments of the X-Ray image.

11. The system of claim 8 , further comprising a report generation module configured to generate a report depicting information about the defect identified in the weld joint, wherein the report is generated upon classifying the X-Ray image in the defective category.

12. The system of claim 8 , wherein the image processing module is further configured to

compute gamma correction values in the X-Ray image by using a mathematical formulation, wherein the mathematical formulation is log(255)/log(pmax−pmin); and

transform the gray scale information of each pixel of the X-Ray image using the standard power law thereby enhancing the contrast of the dark areas and the bright areas in the X-Ray image.

13. The system of claim 8 , wherein the X-Ray is segmented by using at least one of a region-based segmentation and a thresholding technique.

14. The system of claim 8 , wherein the one or more de-noising filters comprises a mean based de-noising, a weighted mean based de-noising, and an adaptive median based de-noising.

15. The system of claim 8 , wherein the analysis module is further configured to compare the gray scale information of the at least one of the dark areas with respect to the one or more neighboring areas to identify the defect in the weld joint.

16. A non transitory computer program product having embodied thereon a computer program carrying one or more instructions for:

capturing an X-Ray image of a weld joint, wherein the X-Ray image comprises dark areas and bright areas, and wherein the dark areas and the bright areas have gray scale information that indicates a luminance intensity;

executing a gamma correction algorithm on the X-Ray image by using a standard power law transform to enhance a contrast of the dark areas and the bright areas, wherein the gamma correction algorithm facilitates a differentiation between the dark areas and bright areas, wherein the gamma correction algorithm is executed on the gray scale information associated with each pixel in the X-Ray image [p min , p max ], wherein p min indicates the gray scale information of a darkest pixel in the X-Ray image, and wherein p max indicates the gray scale information of a brightest pixel in the X-Ray image;

passing, upon execution of the gamma correction algorithm, the X-Ray image through one or more de-noising filters in order to remove noise from the X-Ray image;

identifying, upon removing the noise, a fragment of the X-Ray image that comprises at least one of the dark area with respect to one or more neighboring areas, wherein the at least one of the dark areas is identified by using a gradient discontinuity algorithm, and wherein the at least one of the dark areas indicates a presence of a defect in the weld joint; and

segmenting the X-Ray image into one or more segments based on the identification of the at least one of the dark areas in the X-Ray image, wherein the segmenting facilitates an identification of the defect in the weld joint.

17. The non-transitory computer program product of claim 16 , wherein the X-Ray is segmented by using at least one of a region-based segmentation and a thresholding technique.

18. The non-transitory computer program product of claim 16 , wherein the one or more de-noising filters comprises a mean based de-noising, a weighted mean based de-noising, and an adaptive median based de-noising.

19. The non-transitory computer program product of claim 16 , wherein the computer program further carries instructions for:

comparing the gray scale information of the at least one of the dark areas with respect to the one or more neighboring areas to identify the defect in the weld joint.

20. The non-transitory computer program product of claim 16 , wherein the computer program further carries instructions for:

classifying the weld joint in at least one category of a defective category and a non-defective category.