Apparatus for evaluating quality of video data based on hybrid type and method thereof
View Patent ↗The present invention provides an apparatus for evaluating video quality based on a hybrid type which, when bit stream data is received, extracts first evaluation parameters based on the received bit stream data to calculate a preliminary video quality score based on the first extracted evaluation parameters and then extracts second evaluation parameters, and calculates a final video quality score by adjusting the preliminary video quality score based on the second extracted evaluation parameters and methods thereof.
1. An apparatus for evaluating video quality based on a hybrid type, the apparatus comprising a processor and a non-transitory computer readable medium, wherein the processor executes computer program instructions embodied in the non-transitory computer readable medium to perform the steps of:
when video bit stream data is received, extracting first evaluation parameters from the received video bit stream data;
calculating a preliminary video quality score using a prediction function which is set in advance using the extracted first evaluation parameters;
extracting second evaluation parameters from the received video bit stream data; and
calculating a first final video quality score by adjusting the preliminary video quality score using the extracted second evaluation parameters
wherein the first evaluation parameters comprise a feature value obtained by applying a log function to a bit rate and a feature value obtained by applying a log function to a transmission error rate.
2. The apparatus of claim 1 , wherein the first evaluation parameters comprise a total average QP value, a QP value for I frames, and a log error area value.
3. The apparatus of claim 1 , wherein the second evaluation parameters comprise a feature value obtained by applying a log function to a total number of frozen frames, a feature value obtained by applying a log function to the length of longest frozen frames, and a frame rate.
4. The apparatus of claim 1 , further comprising:
calculating a second final video quality score by secondarily adjusting the first final video quality score using an edge peak signal to noise ratio (EPSNR).
5. The apparatus of claim 1 , further comprising:
calculating a second final video quality score by secondarily adjusting the first final video quality score using an NR video quality score which is obtained by an NR type video quality evaluating method.
6. The apparatus of claim 4 , further comprising:
adjusting the EPSNR by considering video quality reduction due to transmission error degradations before adjusting the first final video quality score using the EPSNR.
7. A computer-implemented method for evaluating video quality based on a hybrid type, comprising:
(a) extracting, when video bit stream data is received, first evaluation parameters which are set in advance based on the received video bit stream data;
(b) calculating a preliminary video quality score using a prediction function which is set in advance using the extracted first evaluation parameters;
(c) extracting second evaluation parameters which are set in advance based on the received video bit stream data; and
(d) calculating a first final video quality score by adjusting the preliminary video quality score using the extracted second evaluation parameters,
wherein the first evaluation parameters comprise a feature value obtained by applying a log function to a bit rate and a feature value obtained by applying a log function to a transmission error rate.
8. The method of claim 7 , wherein the first evaluation parameter comprises a total average QP value, a QP value for I frames, and a log error area value.
9. The method of claim 7 , wherein the second evaluation parameter comprises a feature value obtained by applying a log function to a total number of frozen frames, a feature value obtained by applying a log function to the length of longest frozen frames, and a frame rate.
10. The method of claim 7 , further comprising:
(e) calculating a second final video quality score by secondarily adjusting the first final video quality score using an edge peak signal to noise ratio (EPSNR).
11. The method of claim 7 , further comprising:
(e) calculating a second final video quality score by secondarily adjusting the first final video quality score using an NR video quality score which is obtained by an NR type video quality evaluating method.
12. The method of claim 7 , further comprising:
(e) calculating a second final video quality score by secondarily adjusting the first final video quality score using an adjusted edge peak signal to noise ratio (EPSNR) which is adjusted by considering video quality reduction due to transmission error degradations.