Apparatus and method for predicting compression quality of image in electronic device
A method and/or device for predicting a compression quality of an image during image correction (e.g., image quality enhancement) in an electronic device, and/or processing the image, based on at least the prediction, may be provided. The electronic device may include a display module, a memory, and a processor, wherein the processor may operate to display an image through the display module, extract designated multiple blocks from the image in a designated scheme, estimate confidence for each of the multiple blocks, identify, based on the estimation of the confidence, a first block corresponding to an outlier to be excluded in quality prediction, and a second block for which quality prediction is possible, among the multiple blocks, exclude the first block among the multiple blocks from a subject of quality prediction, and classify a compression quality of the image by using at least the second block remaining after excluding the first block from among the multiple blocks.
1 . An electronic device comprising:
a display;
at least one processor including processing circuitry operatively connected to the display and the memory; and
memory storing instructions that, when executed by the at least one processor individually or collectively, cause the electronic device to:
display compressed image via the display,
extract multiple blocks from the compressed image as a region of interest (ROI),
estimate a confidence value for analyzing a compression rate corresponding to each of the extracted multiple blocks,
identify, based on at least the estimated confidence value, at least one first block having a confidence value equal to or smaller than a designated threshold value,
perform a mean operation for compression rate classification excluding the identified at least one first block from the mean operation, based on at least one second block having a confidence value greater than the designated threshold value, and
determine a compression rate of the compressed image by classifying a result of the mean operation.
2 . The electronic device of claim 1 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to:
select a noise removal model learned to correspond to a compression rate classified for the compressed image,
process quality enhancement for the compressed image, based on the selected noise removal model, to provide an enhanced image, and
process the enhanced image, based on a designated operation.
3 . The electronic device of claim 2 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to:
store, in the memory, multiple noise removal models pre-learned for respective various compression rates, and
select a noise removal model learned to correspond to the classification of the compression rate of the compressed image from among the multiple noise removal models.
4 . The electronic device of claim 2 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to process the enhanced image according to at least one of: displaying through the display, storing in the memory, or transmission outside of the electronic device.
5 . The electronic device of claim 1 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to:
receive a user input of requesting information related to an image quality of the compressed image, and
control, based on the reception of the user input, the display to display correction information of the compressed image, based on the compressed image.
6 . The electronic device of claim 5 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to:
identify at least one block having a compression rate corresponding to a representative quality of the compressed image, based on the reception of the user input, and
provide a designated notification object and detailed information on image correction, based on the identified block.
7 . The electronic device of claim 6 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to:
identify a block having high confidence value from the compressed image,
measure a quality and confidence value for the identified block,
compare each of block-specific compression rates according to the measured quality and confidence value with a representative quality of the compressed image,
perform clustering of blocks each having a compression rate corresponding to the representative quality, among the block-specific compression rates, and having relatively high confidence value compared to other blocks,
provide the designated notification object, based on a part corresponding to the clustered blocks in the compressed image, and
control the display to display detailed information including a total score of the compressed image correction and a description of the classification, together with the notification object.
8 . The electronic device of claim 1 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to, when performing a second compression of the compressed image, remove an artifact generated during a first compression of the compressed image and then perform the second compression.
9 . The electronic device of claim 8 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to:
receive a user input for compression of the compressed image,
based on the reception of the user input, remove the artifact generated during the first compression, based on a noise removal model corresponding to a first compression rate during the first compression of the compressed image,
determine a second compression rate for the second compression, based on the first compression rate related to the first compression, and
perform the second compression of the compressed image, based on the second compression rate.
10 . The electronic device of claim 9 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to determine the second compression rate to correspond to the first compression rate.
11 . The electronic device of claim 1 , wherein the processor comprises a confidence estimation module and a quality prediction module for classification of the compression rate, and
wherein while learning the quality prediction module, the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to learn the quality prediction module, and the confidence estimation module together.
12 . The electronic device of claim 1 , wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to predict the compression rate of the compressed image by using a learning model learned using an artificial intelligence algorithm.
13 . An operation method of an electronic device, the method comprising:
displaying compressed image via a display of the electronic device;
extracting multiple blocks from the compressed image as a region of interest (ROI);
estimating a confidence value for analyzing a compression rate corresponding to each of the extracted multiple blocks;
identifying, based on at least the estimated confidence value, at least one first block having a confidence value equal to or smaller than a designated threshold;
performing a mean operation for the compression rate classification excluding the identified at least one first block from the mean operation, based on at least one second block having a confidence value greater than the designated threshold value; and
determining a compression rate of the compressed image by classifying a result of the mean operation.
14 . The method of claim 13 , further comprising:
storing multiple noise removal models pre-learned for respective various compression rates,
selecting a noise removal model learned to correspond to the classification of the compression rate of the compressed image from among the multiple noise removal models, and
processing quality enhancement for the compressed image, based on the selected noise removal model, to provide an enhanced image.
15 . The method of claim 13 , further comprising:
receiving a user input of requesting information related to an image quality of the compressed image, and
controlling, based on the reception of the user input, to display correction information of the compressed image, based on the compressed image.
16 . The method of claim 15 further comprising:
identifying at least one block having a compression rate corresponding to a representative quality of the compressed image, based on the reception of the user input,
providing a designated notification object, based on a part corresponding to the identified blocks in the compressed image, and
controlling to display detailed information including a total score of the compressed image correction and a description of the classification, together with the notification object.
17 . The method of claim 13 , further comprising:
receiving a user input for compression of the compressed image,
based on the reception of the user input, removing an artifact generated during a first compression, based on a noise removal model corresponding to a first compression rate during the first compression of the compressed image,
determining a second compression rate for a second compression, based on the first compression rate related to the first compression, and
performing the second compression of the compressed image, based on the second compression rate.