Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism
The purpose of the no-reference image quality assessment (NR-IQA) is to measure perceived image quality based on subjective judgments; however, due to the lack of a clean reference image, this is a complicated and unresolved challenge. Massive new IQA datasets have facilitated the creation of deep l...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/4/2682 |
| _version_ | 1797622471286849536 |
|---|---|
| author | Jihyoung Ryu |
| author_facet | Jihyoung Ryu |
| author_sort | Jihyoung Ryu |
| collection | DOAJ |
| description | The purpose of the no-reference image quality assessment (NR-IQA) is to measure perceived image quality based on subjective judgments; however, due to the lack of a clean reference image, this is a complicated and unresolved challenge. Massive new IQA datasets have facilitated the creation of deep learning-based image quality measurements. We present a unique model to handle the NR-IQA challenge in this research by employing a hybrid strategy that leverages from pre-trained CNN model and the unified learning mechanism that extracts both local and non-local characteristics from the input patch. The deep analysis of the proposed framework shows that the model uses features and a mechanism that improves the monotonicity relationship between objective and subjective ratings. The intermediary goal was mapped to a quality score using a regression architecture. To extract various feature maps, a deep architecture with an adaptive receptive field was used. Analyses of this biggest NR-IQA benchmark datasets demonstrate that the suggested technique outperforms current state-of-the-art NR-IQA measures. |
| first_indexed | 2024-03-11T09:10:45Z |
| format | Article |
| id | doaj.art-27f3bc00905d4baeb873934af362b7fd |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T09:10:45Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-27f3bc00905d4baeb873934af362b7fd2023-11-16T18:59:16ZengMDPI AGApplied Sciences2076-34172023-02-01134268210.3390/app13042682Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning MechanismJihyoung Ryu0Electronics and Telecommunications Research Institute (ETRI), Gwangju 61012, Republic of KoreaThe purpose of the no-reference image quality assessment (NR-IQA) is to measure perceived image quality based on subjective judgments; however, due to the lack of a clean reference image, this is a complicated and unresolved challenge. Massive new IQA datasets have facilitated the creation of deep learning-based image quality measurements. We present a unique model to handle the NR-IQA challenge in this research by employing a hybrid strategy that leverages from pre-trained CNN model and the unified learning mechanism that extracts both local and non-local characteristics from the input patch. The deep analysis of the proposed framework shows that the model uses features and a mechanism that improves the monotonicity relationship between objective and subjective ratings. The intermediary goal was mapped to a quality score using a regression architecture. To extract various feature maps, a deep architecture with an adaptive receptive field was used. Analyses of this biggest NR-IQA benchmark datasets demonstrate that the suggested technique outperforms current state-of-the-art NR-IQA measures.https://www.mdpi.com/2076-3417/13/4/2682Inception-ResNet-v2spinal networkimage quality assessment (IQA)no-reference (NR)quality score |
| spellingShingle | Jihyoung Ryu Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism Applied Sciences Inception-ResNet-v2 spinal network image quality assessment (IQA) no-reference (NR) quality score |
| title | Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism |
| title_full | Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism |
| title_fullStr | Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism |
| title_full_unstemmed | Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism |
| title_short | Improved Image Quality Assessment by Utilizing Pre-Trained Architecture Features with Unified Learning Mechanism |
| title_sort | improved image quality assessment by utilizing pre trained architecture features with unified learning mechanism |
| topic | Inception-ResNet-v2 spinal network image quality assessment (IQA) no-reference (NR) quality score |
| url | https://www.mdpi.com/2076-3417/13/4/2682 |
| work_keys_str_mv | AT jihyoungryu improvedimagequalityassessmentbyutilizingpretrainedarchitecturefeatureswithunifiedlearningmechanism |