會議論文
學年 | 110 |
---|---|
學期 | 1 |
發表日期 | 2021-10-12 |
作品名稱 | Effective delivery for block-based compressed sensing with content characteristics |
作品名稱(其他語言) | |
著者 | H.-C. Huang; F.-C. Chang; Y.-Y. Lu; P.-L. Chen |
作品所屬單位 | |
出版者 | |
會議名稱 | 2021 IEEE 10th Global Conference on Consumer Electronics (GCCE) |
會議地點 | Kyoto, Japan |
摘要 | Delivery of compressed images forms an inevitable part in multimedia communications. In addition to conventional compressed schemes in international standards, we employ the newly developed block-based compressed sensing for compression at the encoder, and deliver compressed image over channels. Channel errors may cause degradation to reconstructed qualities at the decoder. We make use of inherent characteristics and color plane correlations of image contents to help enhance the reconstruction quality after delivery. We choose error protection and error regression schemes, and the integration of the two, to look for enhanced reconstruction quality. Experimental results have demonstrated that error protection or regression have revealed their capabilities, and the integration performs the best among different experimental settings. Proposed schemes have demonstrated their advantages to reach enhanced quality of reconstructed images. |
關鍵字 | block-based compressed sensing;error protection;error regression |
語言 | en |
收錄於 | |
會議性質 | 國際 |
校內研討會地點 | 無 |
研討會時間 | 20211012~20211015 |
通訊作者 | |
國別 | JPN |
公開徵稿 | |
出版型式 | |
出處 | 2021 IEEE 10th Global Conference on Consumer Electronics (GCCE), p.820-821 |
相關連結 |
機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/121812 ) |