انتخاب ويژگي براي ارزيابي بي مرجع تصاوير چند تخريبه براساس الگوريتم بهينه سازي ازدحام ذرات

Feature Selection for no reference multi-distortion image quality assessment Based on Particle Swarm Optimization Algorithm

ﺑﺎ ﺗﻮﺟﻪ ﺑﻪ اﻫﻤﯿﺖ ﺗﺼﺎوﯾﺮ در ﮐﺎرﺑﺮدﻫﺎي ﻣﺘﻔﺎوت،ﮐﯿﻔﯿـﺖ آن ﻫـﺎ ﻣـﻮرد ﺗﻮﺟـﻪ اﺳـﺖ. ﻋﻠـﯽ رﻏﻢ ﭘﮋوﻫﺶ ﻫـﺎي اﻧﺠـﺎم ﺷـﺪه در اﯾـﻦ ﺣﻮزه، ﻫﻤﭽﻨﺎن ﻧﻤﯽ ﺗـﻮان ﮐﯿﻔﯿـﺖ ﺗﺼـﺎوﯾﺮ ﺑـﺎ ﺗﺨﺮﯾﺐ ﻫـﺎي ﻣﺘﻔـﺎوت را ﺑـﻪ درﺳـﺘﯽ ﺑﻪ دﺳـﺖ آورد. در اﯾـﻦ ﻣﻘﺎﻟـﻪ، ﯾـﮏ ﻣﻌﯿـﺎر ارزﯾـﺎﺑﯽ ﺑﯽ ﻣﺮﺟﻊ ﮐﯿﻔﯿﺖ ﺗﺼﺎوﯾﺮ ﭼﻨﺪﺗﺨﺮﯾﺒﻪ، ﻣﻌﺮﻓﯽ ﺷﺪه اﺳﺖ. اﯾﻦ ﻣﻌﯿﺎر ﺑﺮﭘﺎﯾﻪ ﺗﺮﮐﯿﺐ وﯾﮋﮔﯽ ﻫﺎي ﺳـﺎﺧﺘﺎري و روﺷـﻨﺎﯾﯽ ﺗﺼـﻮﯾﺮ ، ﮐﯿﻔﯿـﺖ ﺗﺼﺎوﯾﺮ را ﻣﯽ ﺳﻨﺠﺪ. ﻃﺒﻖ ﭘﮋوﻫﺶ ﻫﺎي اﻧﺠﺎم ﺷﺪه، اﯾﻦ وﯾﮋﮔﯽ ﻫﺎ ﺗﺤﺖ ﺗﺎﺛﯿﺮ ﺗﺨﺮﯾﺐ ﻫﺎي ﻣﺘﻔﺎوت، ﺑـﻪ ﺻـﻮرت ﻗﺎﺑـﻞ ﺗـﻮﺟﻬﯽ ﺗﻐﯿﯿـﺮ ﻣﯽ ﮐﻨﻨﺪ. در اﺑﺘﺪا وﯾﮋﮔﯽ ﻫﺎي ﻣﺪﻧﻈﺮ از ﺗﺼﺎوﯾﺮ اﺳﺘﺨﺮاج ﺷﺪﻧﺪ. ﺳﭙﺲ ﺗﻮﺳﻂ اﻟﮕﻮرﯾﺘﻢ ﺑﻬﯿﻨﻪ ﺳﺎزي ازدﺣﺎم ذرات، ﺗﺮﮐﯿـﺐ ﺑﻬﯿﻨـﻪ اي از وﯾﮋﮔﯽ ﻫﺎﯾﯽ ﺑﻪ دﺳﺖ آﻣﺪ ﮐﻪ ﺗﺨﺮﯾﺐ ﻫﺎي ﻣﻮﺟﻮد در ﺗﺼﺎوﯾﺮ ﻣﺠﻤﻮﻋﻪ داده ﻫﺎي ﻣﺘﻔﺎوت را ﺑﻪ درﺳﺘﯽ ﻣﯽ ﺳـﻨﺠﻨﺪ. اﯾـﻦ ﺗﺮﮐﯿـﺐ ، ﺗﻮﺳـﻂ رﮔﺮﺳﯿﻮن ﺑﺮدار ﭘﺸﺘﯿﺒﺎﻧﯽ ﺑﻪ ﻣﺪل آﻣﻮزش داده ﺷﺪ ﺗﺎ ﺑﺘﻮاﻧﺪ ﺑـﻪ ﺑﺮرﺳـﯽ ﮐﯿﻔﯿـﺖ ﺳـﺎﯾﺮ ﺗﺼـﺎوﯾﺮ ﺑـﺎ ﻫﻤـﯿﻦ ﺗﺨﺮﯾﺐ ﻫـﺎ ﺑﭙـﺮدازد. ﺑـﻪ دﻟﯿـﻞ ﺟﺎﻣﻌﯿﺖ وﯾﮋﮔﯽ ﻫﺎي اﻧﺘﺨﺎب ﺷﺪه، اﯾﻦ ﻣﻌﯿـﺎر ﺗﻮاﻧـﺎﯾﯽ ﺳـﻨﺠﺶ ﮐﯿﻔﯿـﺖ ﺗﺼـﺎوﯾﺮ ﺑـﺎ اﻧـﻮاع ﺗﺨﺮﯾﺐ ﻫـﺎ را دارد. ﻃﺒـﻖ ﻧﺘـﺎﯾﺞ ﺣﺎﺻـﻠﻪ از اﺟﺮاي ﻣﻌﯿﺎر، ﺑﻬﺒﻮد ﻗﺎﺑﻞ ﺗﻮﺟﻬﯽ در ارزﯾﺎﺑﯽ ﮐﯿﻔﯿﺖ ﺗﺼﺎوﯾﺮ ﭼﻨﺪ ﺗﺨﺮﯾﺒﻪ و ﺣﺘﯽ ﺗﮏ ﺗﺨﺮﯾﺒﻪ داﺷﺘﯿﻢ. در اﯾﻦ ﭘـﮋوﻫﺶ، ﻋـﻼوه ﺑـﺮ اراﺋـﻪ ﯾﮏ ﻣﻌﯿﺎر ﺟﺎﻣﻊ ﺗﺎﺣﺪ اﻣﮑﺎن ﺑﺘﻮاﻧﺪ اﺑﻌﺎد ﺗﻐﯿﯿﺮ ﯾﺎﻓﺘـﻪ ﺗﺼـﻮﯾﺮ ﺑﻌـﺪ از ﺗﺨﺮﯾـﺐ را ﺑﺴـﻨﺠﺪ، ﺗﺮﮐﯿـﺐ ﺑﻬﯿﻨـﻪ وﯾﮋﮔﯽ ﻫـﺎي ﻣـﻮﺛﺮ در ﺳـﻨﺠﺶ ﮐﯿﻔﯿﺖ ﺗﺼﺎوﯾﺮ ﺗﺤﺖ ﺗﺎﺛﯿﺮ ﺗﺨﺮﯾﺐ ﯾﺎ ﺗﺨﺮﯾﺐ ﻫﺎي ﻣﺘﻔﺎوت،ﺑﻪ دﺳﺖ آﻣﺪ.

In this paper, a no-reference metric for evaluating the quality of multi-distortion images is introduced. This metric is based on a combination of structural features and image brightness. First, the structural features and brightness of the image, which change drastically due to distortion, were extracted. For different datasets, an optimal combination of properties was obtained by the particle swarm optimization algorithm. The optimal combination of features was supported by regression vector regression to the training model so that the trained model could measure the quality of other images. Due to the comprehensiveness of the selected features, this metric has the ability to measure image quality with a variety of degradations. According to the results of the implementation of the criterion, we had a significant improvement and also according to the research, the optimal combination of image properties has been obtained to investigate specific degradations, which can be useful for further research in the future.