WEKO3
アイテム
{"_buckets": {"deposit": "a28bfad8-e8d5-4284-b0dc-732fc85fba7e"}, "_deposit": {"created_by": 15, "id": "10321", "owners": [15], "pid": {"revision_id": 0, "type": "depid", "value": "10321"}, "status": "published"}, "_oai": {"id": "oai:sucra.repo.nii.ac.jp:00010321", "sets": ["504"]}, "author_link": [], "item_113_alternative_title_1": {"attribute_name": "タイトル(別言語)", "attribute_value_mlt": [{"subitem_alternative_title": "ブラインド条件下での雑音分散推定に基づく雑音付加ボケ画像復元"}]}, "item_113_biblio_info_9": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2014", "bibliographicIssueDateType": "Issued"}}]}, "item_113_date_35": {"attribute_name": "作成日", "attribute_value_mlt": [{"subitem_date_issued_datetime": "2015-02-04", "subitem_date_issued_type": "Created"}]}, "item_113_date_granted_20": {"attribute_name": "学位授与年月日", "attribute_value_mlt": [{"subitem_dategranted": "2014-03-24"}]}, "item_113_degree_grantor_22": {"attribute_name": "学位授与機関", "attribute_value_mlt": [{"subitem_degreegrantor": [{"subitem_degreegrantor_name": "埼玉大学"}], "subitem_degreegrantor_identifier": [{"subitem_degreegrantor_identifier_name": "12401", "subitem_degreegrantor_identifier_scheme": "kakenhi"}]}]}, "item_113_degree_name_21": {"attribute_name": "学位名", "attribute_value_mlt": [{"subitem_degreename": "博士(工学)"}]}, "item_113_description_13": {"attribute_name": "形態", "attribute_value_mlt": [{"subitem_description": "82 p.", "subitem_description_type": "Other"}]}, "item_113_description_23": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "Many image restoration algorithms have been proposed over the last few decades. But in early work on image restoration, it was nearly always assumed that all the information required to restore an image is known previously. Unfortunately, this is not possible in most real-life situations. Thus, the technique of blind image restoration estimating both the true image and the blur from a degraded image has been researched. However, when taking noise into account, the estimation problem becomes more challenging. Estimating the blue and noise parameters simultaneously may cause a large estimation error.\n To minimize the estimation error, we propose a Maximum-Likelihood Estimation Algorithm based on noise variance estimation. This method estimates the noise variance only using the information from the known degraded image. This improves the estimation accuracy significantly.\n For better results for noise variance estimation from the degraded image, we propose a structure-based method. This method separates an image into blocks. Rejecting the edge included blocks, only the homogenous blocks are selected for the noise estimation process. Since image details can be better revealed by second-order operators, for better estimation, we further propose a difference eigenvalue edge indicator with threshold for more accurate block selection.\n After combining the noise estimation method, better estimation results of blur and noise are derived, leading to the better restoration results. Experiments show that the structure-based method is good for light noisy condition, while the difference eigenvalue based method is effective for fine texture images.", "subitem_description_type": "Abstract"}]}, "item_113_description_24": {"attribute_name": "目次", "attribute_value_mlt": [{"subitem_description": "Acknowledgements .............................................................................................................. 11\nAbstract................................................................................................................................ 12\n1 Introduction ................................................................................................................... 13\n1.1 Problem Formulation.......................................................................................... 15\n1.2 Important Characteristics of Blind Deconvolution .......................................... 16\n1.3 Existing Approaches in Blind Deconvolution ................................................... 17\n1.3.1 Zero Sheet Separation .............................................................................. 17\n1.3.2 A Priori Blur Identification Methods ...................................................... 17\n1.3.3 ARMA Parameter Estimation Methods .................................................. 18\n1.3.4 Nonparametric Deterministic Image Constraints Restoration Techniques................................................................................................ 18\n1.3.5 Nonparametric Methods Based on High Order Statistics ..................... 18\n1.4 Related Works ..................................................................................................... 19\n1.5 Thesis Organization............................................................................................ 19\n2 Improved Maximum-Likelihood Estimation Algorithm for Blind Image Deconvolution Based on Noise Variance Estimation .................................................. 20\n2.1 Related Works ..................................................................................................... 21\n2.2 Image Model and Estimation ............................................................................. 22\n2.2.1 ARMA model ............................................................................................. 22\n2.2.2 ML Estimation Algorithm ........................................................................ 23\n2.2.3 EM Algorithm............................................................................................ 24\n2.3 Blind Image Deconvolution Based on Noise Variance Estimation.................. 26\n2.3.1 Estimating the Noise Variance from the Observed Image ........................................................................................................ 26\n2.3.2 Estimating Blur Parameters for Different Blur Types .......................... 27\n2.4 Experimental Results ......................................................................................... 30\n2.4.1 Linear Motion Blur ................................................................................... 31\n2.4.2 Uniform Defocus Blur............................................................................... 32\n2.4.3 Gaussian Blur ........................................................................................... 34\n2.4.4 Other Blur ................................................................................................. 37\n2.5 Summary ............................................................................................................. 38\n3 Improved Structure-Based Gaussian Noise Variance Estimation Method .............. 39\n3.1 Related Works ..................................................................................................... 39\n3.2 Proposed Image Noise Variance Estimation Method ....................................... 41\n3.2.1 Homogeneity Measure.............................................................................. 41\n3.2.2 Threshold for Variance Averaging ........................................................... 45\n3.3 Experimental Results ......................................................................................... 47\n3.3.1 Comparison of Masks ............................................................................... 50\n3.3.2 Comparison of Threshold ......................................................................... 50\n3.3.3 Combining the masks and threshold ...................................................... 50\n3.3.4 Comparison of Images with Homogeneous Areas and Images with Fine Texture ............................................................................................. 51\n3.4 Summary ............................................................................................................. 57\n4 Difference Eigenvalue Based Gaussian Noise Variance Estimation Method........... 58\n4.1 Related Works ..................................................................................................... 58\n4.2 Proposed Image Noise Variance Estimation Method ....................................... 59\n4.2.1 Hessian Matrix ......................................................................................... 60\n4.2.2 Difference Eigenvalue Edge Indicator .................................................... 61\n4.2.3 Threshold for Selection of Homogeneous Blocks .................................... 61\n4.3 Experimental Results ......................................................................................... 62\n4.3.1 Results of Homogeneous Pixels Selection ............................................... 64\n4.3.2 Comparison Results.................................................................................. 68\n4.4 Summary ............................................................................................................. 68\n5 Conclusions .................................................................................................................... 72\n5.1 Summary of the Research .................................................................................. 72\n5.2 Future Works....................................................................................................... 73", "subitem_description_type": "Other"}]}, "item_113_description_25": {"attribute_name": "注記", "attribute_value_mlt": [{"subitem_description": "主指導教員 : 島村徹也", "subitem_description_type": "Other"}]}, "item_113_description_33": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"subitem_description": "text", "subitem_description_type": "Other"}]}, "item_113_description_34": {"attribute_name": "フォーマット", "attribute_value_mlt": [{"subitem_description": "application/pdf", "subitem_description_type": "Other"}]}, "item_113_dissertation_number_19": {"attribute_name": "学位授与番号", "attribute_value_mlt": [{"subitem_dissertationnumber": "甲第948号"}]}, "item_113_identifier_registration": {"attribute_name": "ID登録", "attribute_value_mlt": [{"subitem_identifier_reg_text": "10.24561/00010315", "subitem_identifier_reg_type": "JaLC"}]}, "item_113_publisher_11": {"attribute_name": "出版者名", "attribute_value_mlt": [{"subitem_publisher": "埼玉大学大学院理工学研究科"}]}, "item_113_publisher_12": {"attribute_name": "出版者名(別言語)", "attribute_value_mlt": [{"subitem_publisher": "Graduate School of Science and Engineering, Saitama University"}]}, "item_113_record_name_8": {"attribute_name": "書誌", "attribute_value_mlt": [{"subitem_record_name": "博士論文(埼玉大学大学院理工学研究科(博士後期課程))"}]}, "item_113_text_31": {"attribute_name": "版", "attribute_value_mlt": [{"subitem_text_value": "[出版社版]"}]}, "item_113_text_36": {"attribute_name": "アイテムID", "attribute_value_mlt": [{"subitem_text_value": "GD0000538"}]}, "item_113_text_4": {"attribute_name": "著者 所属", "attribute_value_mlt": [{"subitem_text_value": "埼玉大学大学院理工学研究科(博士後期課程)"}]}, "item_113_text_5": {"attribute_name": "著者 所属(別言語)", "attribute_value_mlt": [{"subitem_text_value": "Graduate School of Science and Engineering, Saitama University"}]}, "item_113_version_type_32": {"attribute_name": "著者版フラグ", "attribute_value_mlt": [{"subitem_version_resource": "http://purl.org/coar/version/c_970fb48d4fbd8a85", "subitem_version_type": "VoR"}]}, "item_access_right": {"attribute_name": "アクセス権", "attribute_value_mlt": [{"subitem_access_right": "open access", "subitem_access_right_uri": "http://purl.org/coar/access_right/c_abf2"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Yi, Chong", "creatorNameLang": "en"}, {"creatorName": "イ, チョン", "creatorNameLang": "ja-Kana"}]}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2018-01-23"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "GD0000538.pdf", "filesize": [{"value": "2.4 MB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_note", "mimetype": "application/pdf", "size": 2400000.0, "url": {"label": "GD0000538.pdf", "objectType": "fulltext", "url": "https://sucra.repo.nii.ac.jp/record/10321/files/GD0000538.pdf"}, "version_id": "53910914-26a9-48e3-9015-05d75fb74103"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "eng"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "doctoral thesis", "resourceuri": "http://purl.org/coar/resource_type/c_db06"}]}, "item_title": "Noisy Blurred Image Restoration Based on Noise Variance Estimation in Blind Condition", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Noisy Blurred Image Restoration Based on Noise Variance Estimation in Blind Condition", "subitem_title_language": "en"}]}, "item_type_id": "113", "owner": "15", "path": ["504"], "permalink_uri": "https://doi.org/10.24561/00010315", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2015-02-04"}, "publish_date": "2015-02-04", "publish_status": "0", "recid": "10321", "relation": {}, "relation_version_is_last": true, "title": ["Noisy Blurred Image Restoration Based on Noise Variance Estimation in Blind Condition"], "weko_shared_id": -1}
Noisy Blurred Image Restoration Based on Noise Variance Estimation in Blind Condition
https://doi.org/10.24561/00010315
https://doi.org/10.24561/00010315aedb9e87-d8ca-4812-beb1-1a8b6eab9676
名前 / ファイル | ライセンス | アクション |
---|---|---|
GD0000538.pdf (2.4 MB)
|
|
Item type | 学位論文 / Thesis or Dissertation(1) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
公開日 | 2015-02-04 | |||||||||
タイトル | ||||||||||
言語 | en | |||||||||
タイトル | Noisy Blurred Image Restoration Based on Noise Variance Estimation in Blind Condition | |||||||||
言語 | ||||||||||
言語 | eng | |||||||||
資源タイプ | ||||||||||
資源タイプ識別子 | http://purl.org/coar/resource_type/c_db06 | |||||||||
資源タイプ | doctoral thesis | |||||||||
ID登録 | ||||||||||
ID登録 | 10.24561/00010315 | |||||||||
ID登録タイプ | JaLC | |||||||||
アクセス権 | ||||||||||
アクセス権 | open access | |||||||||
アクセス権URI | http://purl.org/coar/access_right/c_abf2 | |||||||||
タイトル(別言語) | ||||||||||
その他のタイトル | ブラインド条件下での雑音分散推定に基づく雑音付加ボケ画像復元 | |||||||||
著者 |
Yi, Chong
× Yi, Chong
|
|||||||||
著者 所属 | ||||||||||
埼玉大学大学院理工学研究科(博士後期課程) | ||||||||||
著者 所属(別言語) | ||||||||||
Graduate School of Science and Engineering, Saitama University | ||||||||||
書誌 | ||||||||||
収録物名 | 博士論文(埼玉大学大学院理工学研究科(博士後期課程)) | |||||||||
書誌情報 |
発行日 2014 |
|||||||||
出版者名 | ||||||||||
出版者 | 埼玉大学大学院理工学研究科 | |||||||||
出版者名(別言語) | ||||||||||
出版者 | Graduate School of Science and Engineering, Saitama University | |||||||||
形態 | ||||||||||
内容記述タイプ | Other | |||||||||
内容記述 | 82 p. | |||||||||
学位授与番号 | ||||||||||
学位授与番号 | 甲第948号 | |||||||||
学位授与年月日 | ||||||||||
学位授与年月日 | 2014-03-24 | |||||||||
学位名 | ||||||||||
学位名 | 博士(工学) | |||||||||
学位授与機関 | ||||||||||
学位授与機関識別子Scheme | kakenhi | |||||||||
学位授与機関識別子 | 12401 | |||||||||
学位授与機関名 | 埼玉大学 | |||||||||
抄録 | ||||||||||
内容記述タイプ | Abstract | |||||||||
内容記述 | Many image restoration algorithms have been proposed over the last few decades. But in early work on image restoration, it was nearly always assumed that all the information required to restore an image is known previously. Unfortunately, this is not possible in most real-life situations. Thus, the technique of blind image restoration estimating both the true image and the blur from a degraded image has been researched. However, when taking noise into account, the estimation problem becomes more challenging. Estimating the blue and noise parameters simultaneously may cause a large estimation error. To minimize the estimation error, we propose a Maximum-Likelihood Estimation Algorithm based on noise variance estimation. This method estimates the noise variance only using the information from the known degraded image. This improves the estimation accuracy significantly. For better results for noise variance estimation from the degraded image, we propose a structure-based method. This method separates an image into blocks. Rejecting the edge included blocks, only the homogenous blocks are selected for the noise estimation process. Since image details can be better revealed by second-order operators, for better estimation, we further propose a difference eigenvalue edge indicator with threshold for more accurate block selection. After combining the noise estimation method, better estimation results of blur and noise are derived, leading to the better restoration results. Experiments show that the structure-based method is good for light noisy condition, while the difference eigenvalue based method is effective for fine texture images. |
|||||||||
目次 | ||||||||||
内容記述タイプ | Other | |||||||||
内容記述 | Acknowledgements .............................................................................................................. 11 Abstract................................................................................................................................ 12 1 Introduction ................................................................................................................... 13 1.1 Problem Formulation.......................................................................................... 15 1.2 Important Characteristics of Blind Deconvolution .......................................... 16 1.3 Existing Approaches in Blind Deconvolution ................................................... 17 1.3.1 Zero Sheet Separation .............................................................................. 17 1.3.2 A Priori Blur Identification Methods ...................................................... 17 1.3.3 ARMA Parameter Estimation Methods .................................................. 18 1.3.4 Nonparametric Deterministic Image Constraints Restoration Techniques................................................................................................ 18 1.3.5 Nonparametric Methods Based on High Order Statistics ..................... 18 1.4 Related Works ..................................................................................................... 19 1.5 Thesis Organization............................................................................................ 19 2 Improved Maximum-Likelihood Estimation Algorithm for Blind Image Deconvolution Based on Noise Variance Estimation .................................................. 20 2.1 Related Works ..................................................................................................... 21 2.2 Image Model and Estimation ............................................................................. 22 2.2.1 ARMA model ............................................................................................. 22 2.2.2 ML Estimation Algorithm ........................................................................ 23 2.2.3 EM Algorithm............................................................................................ 24 2.3 Blind Image Deconvolution Based on Noise Variance Estimation.................. 26 2.3.1 Estimating the Noise Variance from the Observed Image ........................................................................................................ 26 2.3.2 Estimating Blur Parameters for Different Blur Types .......................... 27 2.4 Experimental Results ......................................................................................... 30 2.4.1 Linear Motion Blur ................................................................................... 31 2.4.2 Uniform Defocus Blur............................................................................... 32 2.4.3 Gaussian Blur ........................................................................................... 34 2.4.4 Other Blur ................................................................................................. 37 2.5 Summary ............................................................................................................. 38 3 Improved Structure-Based Gaussian Noise Variance Estimation Method .............. 39 3.1 Related Works ..................................................................................................... 39 3.2 Proposed Image Noise Variance Estimation Method ....................................... 41 3.2.1 Homogeneity Measure.............................................................................. 41 3.2.2 Threshold for Variance Averaging ........................................................... 45 3.3 Experimental Results ......................................................................................... 47 3.3.1 Comparison of Masks ............................................................................... 50 3.3.2 Comparison of Threshold ......................................................................... 50 3.3.3 Combining the masks and threshold ...................................................... 50 3.3.4 Comparison of Images with Homogeneous Areas and Images with Fine Texture ............................................................................................. 51 3.4 Summary ............................................................................................................. 57 4 Difference Eigenvalue Based Gaussian Noise Variance Estimation Method........... 58 4.1 Related Works ..................................................................................................... 58 4.2 Proposed Image Noise Variance Estimation Method ....................................... 59 4.2.1 Hessian Matrix ......................................................................................... 60 4.2.2 Difference Eigenvalue Edge Indicator .................................................... 61 4.2.3 Threshold for Selection of Homogeneous Blocks .................................... 61 4.3 Experimental Results ......................................................................................... 62 4.3.1 Results of Homogeneous Pixels Selection ............................................... 64 4.3.2 Comparison Results.................................................................................. 68 4.4 Summary ............................................................................................................. 68 5 Conclusions .................................................................................................................... 72 5.1 Summary of the Research .................................................................................. 72 5.2 Future Works....................................................................................................... 73 |
|||||||||
注記 | ||||||||||
内容記述タイプ | Other | |||||||||
内容記述 | 主指導教員 : 島村徹也 | |||||||||
版 | ||||||||||
[出版社版] | ||||||||||
著者版フラグ | ||||||||||
出版タイプ | VoR | |||||||||
出版タイプResource | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |||||||||
資源タイプ | ||||||||||
内容記述タイプ | Other | |||||||||
内容記述 | text | |||||||||
フォーマット | ||||||||||
内容記述タイプ | Other | |||||||||
内容記述 | application/pdf | |||||||||
作成日 | ||||||||||
日付 | 2015-02-04 | |||||||||
日付タイプ | Created | |||||||||
アイテムID | ||||||||||
GD0000538 |