Image recognition method and image recognition apparatus

• US 20090010547A1
• Filed: 12/06/2007
• Published: 01/08/2009
• Est. Priority Date: 07/06/2007
• Status: Active Grant

• Current Assignee: Quanta Computer
• Original Assignee: Quanta Computer
• Inventors: Chiang, Jia-Jie, Lin, Cheng-Wei, Liao, Po-Hsuan
• Granted Patent: US 8,064,707 B2
• Time in Patent Office: Days
• US Class Current: 382/207
• CPC Class Codes: G06V 10/431 Frequency domain transformation; Autocorrelation
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis

Abstract

The invention provides an image recognition method. First, it is judged whether a set of discrete cosine transform (DCT) coefficients corresponding to an image and/or a set of texture parameters corresponding to the DCT coefficients exist. If the judgment is no, the image is selectively performed a DCT or an inverse discrete cosine transform (IDCT) to generate the set of DCT coefficients based on a format of the image. Based on the set of DCT coefficients, the set of texture parameters is then generated. Afterward, the set of texture parameters is compared with a set of target texture parameters to generate a recognition result.

Claims

1. An image recognition method, comprising the steps of:

(a) judging whether a set of discrete cosine transform (DCT) coefficients corresponding to an image and/or a set of texture parameters corresponding to the set of DCT coefficients exist;

(b) if the judgment of step (a) is no, generating the set of DCT coefficients by selectively performing a DCT or an inverse discrete cosine transform (IDCT) to the image according to a format of the image, and generating the set of texture parameters according to the set of DCT coefficients; and

(c) comparing the set of texture parameters with a set of target texture parameters to generate a recognition result.

2. The method of claim 1, further comprising the step of:

(d) if the judgment of step (a) is that the set of texture parameters has existed, comparing the set of texture parameters with the set of target texture parameters to generate the recognition result.

3. The method of claim 1, further comprising the step of:

(e) if the judgment of step (a) is that the set of DCT coefficients has existed, but the set of texture parameters does not exist, generating the set of texture parameters according to the set of DCT coefficients, and comparing the set of texture parameters with the set of target texture parameters to generate the recognition result.

4. The method of claim 1, wherein if the format is a BMP format, the DCT is performed to the image in step (b).

5. The method of claim 4, further comprising the step of:

(f) before the DCT is performed, performing a grey level transform to the image.

6. The method of claim 1, wherein if the format is a JPEG format, the IDCT is performed to the image in step (b).

7. The method of claim 1, wherein the set of texture parameters comprises a first smooth texture energy (E_DC1), a first vertical/horizontal texture energy ratio [(E_V1/E_H1)] and a first oblique texture energy (E_S1), the set of target texture parameters comprises a target smooth texture energy (E_DC), a target vertical/horizontal texture energy ratio [(E_V/E_H)] and a target oblique texture energy (E_S), and in step (c), the recognition result is generated according to E_DC1, (E_V1/E_H1), E_S1, E_DC, E_V/E_H, and E_S.

8. The method of claim 7, wherein the recognition result (D) can be represented as:

$D=a\times <br><br><br><br>E_{D<br><br><br><br>C<br><br><br><br>1}-E_{D<br><br><br><br>C}<br><br>+b\times <br><br><br><br>\frac{E_{V<br><br><br><br>1}}{E_{H<br><br><br><br>1}}-\frac{E_V}{E_H}<br><br>+c\times <br><br><br><br>E_{S<br><br><br><br>1}-E_S<br><br>,$

wherein a, b, and c are weighing coefficients.

9. The method of claim 1, wherein the set of target texture parameters corresponds to a target image, and the recognition result is a degree of similarity between the image and the target image.

10. The method of claim 9, wherein the image is among the N candidate images, N is a positive integer, and each of the N candidate images corresponds to a candidate recognition result respectively, the method further comprising the step of:

(g) selecting a resulting image which is most similar to the target image from the N candidate images based on the recognition results of the candidates.

11. The method of claim 1, further comprising the step of:

(h) after step (b) is performed, storing the set of DCT coefficients and/or the set of texture parameters.

12. An image recognition apparatus, comprising:

a judging module, for judging whether a set of discrete cosine transform (DCT) coefficients corresponding to an image and/or a set of texture parameters corresponding to the set of DCT coefficients exist;

a DCT/IDCT module, electrically connected to the judging module, if the judgment of the judging module is no, generating the set of DCT coefficients by selectively performing a DCT or an inverse discrete cosine transform (IDCT) to the image according to a format of the image, and generating the set of texture parameters according to the set of DCT coefficients; and

a comparing module, electrically connected to the DCT/IDCT module, for comparing the set of texture parameters with a set of target texture parameters to generate a recognition result.

13. The apparatus of claim 12, wherein if the judgment of the judging module is that the set of texture parameters has existed, the comparing module compares the set of texture parameters with the set of target texture parameters to generate the recognition result.

14. The apparatus of claim 12, wherein if the judgment of the judging module is that the set of DCT coefficients has existed, but the set of texture parameters does not exist, the DCT/IDCT module generates the set of texture parameters according to the set of DCT coefficients, and the comparing module compares the set of texture parameters with the set of target texture parameters to generate the recognition result.

15. The apparatus of claim 12, wherein if the format is a BMP format, the DCT is performed to the image by the DCT/IDCT module.

16. The apparatus of claim 14, further comprising:

a grey level transform module, electrically connected to the DCT/IDCT module, for performing a grey level transform to the image before the DCT is performed by the DCT/IDCT module.

17. The apparatus of claim 12, wherein if the format is a JPEG format, the IDCT is performed to the image by the DCT/IDCT module.

18. The apparatus of claim 12, wherein the set of texture parameters comprise a first smooth texture energy (E_DC1), a first vertical/horizontal texture energy ratio [(E_V1/E_H1)] and a first oblique texture energy (E_S1);

the set of target texture parameters comprises a target smooth texture energy (E_DC), a target vertical/horizontal texture energy ratio [(E_V/E_H)] and a target oblique texture energy (E_S), and the recognition result is generated by the comparing module according to E_DC1, (E_V1/E_H1), E_S1, E_DC, E_V/E_H, and E_S.

19. The apparatus of claim 18, wherein the recognition result (D) can be represented as:

$D=a\times <br><br><br><br>E_{D<br><br><br><br>C<br><br><br><br>1}-E_{D<br><br><br><br>C}<br><br>+b\times <br><br><br><br>\frac{E_{V<br><br><br><br>1}}{E_{H<br><br><br><br>1}}-\frac{E_V}{E_H}<br><br>+c\times <br><br><br><br>E_{S<br><br><br><br>1}-E_S<br><br>,$

wherein a, b, and c are weighing coefficients.

20. The apparatus of claim 12, wherein the set of target texture parameters corresponds to a target image, and the recognition result is a degree of similarity between the image and the target image.

21. The apparatus of claim 20, wherein N candidate images comprise the image, N is a positive integer, and each of the N candidate images corresponds to a candidate recognition result respectively, the apparatus further comprising:

a selecting module, electrically connected to the comparing module, for selecting a resulting image which is most similar to the target image from the N candidate images according to the candidate recognition results.

22. The apparatus of claim 12, further comprising:

a storing module, electrically connected to the DCT/IDCT module, for storing the set of DCT coefficients and/or the set of texture parameters.