• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Multimedia Forensics for Classification of Fake Content

Research Project

Project/Area Number 19K22846
Research Category

Grant-in-Aid for Challenging Research (Exploratory)

Allocation TypeMulti-year Fund
Review Section Medium-sized Section 60:Information science, computer engineering, and related fields
Research InstitutionOkayama University

Principal Investigator

Kuribayashi Minoru  岡山大学, 自然科学学域, 准教授 (50346235)

Project Period (FY) 2019-06-28 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥6,370,000 (Direct Cost: ¥4,900,000、Indirect Cost: ¥1,470,000)
Fiscal Year 2021: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2020: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2019: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Keywordsフェイクコンテンツ / 生成AI / 画像識別器 / 敵対的サンプル / 敵対的生成ネットワーク / ディープフェイク / 敵対的攻撃 / 敵対的事例 / 改ざん検知 / ノイズ除去フィルタ / 真偽判定 / 畳み込みニューラルネットワーク / 深層学習 / 信号処理 / マルチメディアフォレンジクス
Outline of Research at the Start

マルチメディアコンテンツの真偽判定技術を実現させるため,本研究では「人工的に創造された画像」と「正常に撮影された画像」を分類する手法を開発する.人工的に創造される過程において生じる不自然な信号成分を検出して,分類精度を高めるために,画像データベースの画像に対してフェイク画像を作成し,それぞれのラベル付き教師データを準備して識別器に学習させる.また,敵対的に雑音を加えることで識別不能にする攻撃に対応するために,フィルタ操作による信号処理の影響を解析して,攻撃が行われたか否かを判定するシステムを開発する予定である.

Outline of Final Research Achievements

We designed and evaluated a basic image classifier for fake content created by replacing a person's face with another person's face or changing facial expressions, as well as for images and videos artificially created by an adversarial generative network. We also proposed extensions to the region-based processing and color space correction methods to achieve high classification accuracy for content created with different methods during training and testing. In addition, we developed a method to identify the presence or absence of adversarial noise generated to misidentify the image classifier. We designed a preprocessing filter that does not modify the original image classifier and confirmed through experiments that it can detect adversarial images with a high accuracy of over 90%.

Academic Significance and Societal Importance of the Research Achievements

深層学習技術を用いれば,偽物とは簡単に判別できない画像や映像を生成することが可能となっており,悪用されることが懸念されている.特に,特定の人物を操った形でのコンテンツが精巧に作成され,社会的な影響を与えるような印象操作に使われる可能性も高まっている.本研究で開発した識別器は,そのような人工的に作成および加工編集されたコンテンツを機械的に判別するものであり,マルチメディアコンテンツに関連したセキュリティ技術として重要な役割を担うものと考えられる.本識別器を拡張させて更に精度を高める研究が広がることで,フェイク情報の流布を抑制できることが期待される.

Report

(5 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • 2019 Research-status Report
  • Research Products

    (65 results)

All 2023 2022 2021 2020 2019 Other

All Int'l Joint Research (10 results) Journal Article (9 results) (of which Int'l Joint Research: 7 results,  Peer Reviewed: 9 results,  Open Access: 5 results) Presentation (45 results) (of which Int'l Joint Research: 20 results,  Invited: 2 results) Book (1 results)

  • [Int'l Joint Research] Ahmedabad University(インド)

    • Related Report
      2022 Annual Research Report
  • [Int'l Joint Research] Aligarh Muslim University(インド)

    • Related Report
      2022 Annual Research Report
  • [Int'l Joint Research] カタルーニャ・オベルタ大学(スペイン)

    • Related Report
      2021 Research-status Report
  • [Int'l Joint Research] ワルシャワ工科大学(ポーランド)

    • Related Report
      2021 Research-status Report
  • [Int'l Joint Research] モナーシュ大学マレーシア(マレーシア)

    • Related Report
      2021 Research-status Report
  • [Int'l Joint Research] アーメダバード大学(インド)

    • Related Report
      2021 Research-status Report
  • [Int'l Joint Research] カタルーニャ・オベルタ大学(スペイン)

    • Related Report
      2020 Research-status Report
  • [Int'l Joint Research] ワルシャワ工科大学(ポーランド)

    • Related Report
      2020 Research-status Report
  • [Int'l Joint Research] モナーシュ大学マレーシア(マレーシア)

    • Related Report
      2020 Research-status Report
  • [Int'l Joint Research] アーメダバード大学(インド)

    • Related Report
      2020 Research-status Report
  • [Journal Article] Effects of Image Processing Operations on Adversarial Noise and Their Use in Detecting and Correcting Adversarial Images2022

    • Author(s)
      Huy H. Nguyen, Minoru Kuribayashi, Junichi Yamagishi, Isao Echizen
    • Journal Title

      IEICE Transactions on Information and Systems

      Volume: E105.D Issue: 1 Pages: 65-77

    • DOI

      10.1587/transinf.2021MUP0005

    • NAID

      130008138838

    • ISSN
      0916-8532, 1745-1361
    • Year and Date
      2022-01-01
    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Coded DNN watermark: robustness against pruning models using constant weight code2022

    • Author(s)
      T. Yasui, T. Tanaka, A. Malik, M. Kuribayashi
    • Journal Title

      Journal of Imaging

      Volume: vol.8, no.6 Issue: 6 Pages: 152-152

    • DOI

      10.3390/jimaging8060152

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Classification of Screenshot Image Captured in Online Meeting System2022

    • Author(s)
      M. Kuribayashi, K. Kamakari, N. Funabiki
    • Journal Title

      Lecture Notes in Computer Science

      Volume: 13480 Pages: 244-255

    • DOI

      10.1007/978-3-031-14463-9_16

    • ISBN
      9783031144622, 9783031144639
    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] DeepFake Detection for Human Face Images and Videos: A Survey2022

    • Author(s)
      Asad Malik, Minoru Kuribayashi, Sani M. Abdullahi, Ahmad Neyaz Khan
    • Journal Title

      IEEE Access

      Volume: 10 Pages: 18757-18775

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Architecture of a fake news detection system combining digital watermarking, signal processing, and machine learning2022

    • Author(s)
      D. Megias, M. Kuribayashi, A. Rosales, K. Cabaj, W. Mazurczyk
    • Journal Title

      Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications

      Volume: 13 Pages: 33-55

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] StealthPDF: Data hiding method for PDF file with no visual degradation2021

    • Author(s)
      Minoru Kuribayashi, KokSheik Wong
    • Journal Title

      Journal of Information Security and Applications

      Volume: 61 Pages: 10-10

    • DOI

      10.1016/j.jisa.2021.102875

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Survey on Decentralized Fingerprinting Solutions: Copyright Protection through Piracy Tracing2020

    • Author(s)
      Megias David, Kuribayashi Minoru, Qureshi Amna
    • Journal Title

      Computers

      Volume: 9 Issue: 2 Pages: 26-26

    • DOI

      10.3390/computers9020026

    • Related Report
      2020 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Improved DM-QIM Watermarking Scheme for PDF Document2020

    • Author(s)
      Minoru Kuribayashi and KokSheik Wong
    • Journal Title

      Digital Forensics and Watermarking

      Volume: 12022 Pages: 171-183

    • DOI

      10.1007/978-3-030-43575-2_14

    • ISBN
      9783030435745, 9783030435752
    • Related Report
      2019 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Near-optimal detector for binary tardos code by estimating collusion strategy2019

    • Author(s)
      Tatsuya Yasui, Minoru Kuribayashi, Nobuo Funabiki, and Isao Echizen
    • Journal Title

      IEEE Transactions on Information Forensics and Security

      Volume: 15 Pages: 2069-2080

    • DOI

      10.1109/tifs.2019.2956587

    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Presentation] プライバシ保護に向けたOpenPose骨格推定を用いた人の頭部位置検出2023

    • Author(s)
      瀬尾亮太, 田中千奈月, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 3月
    • Related Report
      2022 Annual Research Report
  • [Presentation] OpenPoseで検出した特徴点における動作特性に注目した歩容認証2023

    • Author(s)
      田中千奈月, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 3月
    • Related Report
      2022 Annual Research Report
  • [Presentation] 赤井怜音, 栗林稔, 舩曵信生2023

    • Author(s)
      Federated Learningにおいて悪意のある分散ノードを除外するための一検討
    • Organizer
      信学技報, EMM 3月
    • Related Report
      2022 Annual Research Report
  • [Presentation] 敵対的生成ネットワークによって生成されたシンセティックメディア識別のための画像処理解析の検討2023

    • Author(s)
      浦晃暢, 栗林稔, 舩曵信生
    • Organizer
      2023年暗号と情報セキュリティシンポジウム (SCIS2023)
    • Related Report
      2022 Annual Research Report
  • [Presentation] ブロックチェーン技術によるコンテンツの加工編集履歴を管理するコンテンツクレデンシャル機能の一考案2023

    • Author(s)
      笠井健太郎, 栗林稔, 舩曵信生, 越前功
    • Organizer
      2023年暗号と情報セキュリティシンポジウム (SCIS2023)
    • Related Report
      2022 Annual Research Report
  • [Presentation] Multimedia Forensics for Fake Content Classification2023

    • Author(s)
      Minoru Kuribayashi
    • Organizer
      International Conference on Emerging Computational Intelligence (ICECI 2023)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] A Vision Transformer-Based Approach to Bearing Fault Classification via Vibration Signals2022

    • Author(s)
      A. H. Zim, A. Ashraf, A. Iqbal, A. Malik, M. Kuribayashi
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Reversible Data Hiding in Encrypted Text Using Paillier Cryptosystem2022

    • Author(s)
      A. Malik, A. Ashraf, H. Wu, M. Kuribayashi
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Detection and Correction of Adversarial Examples Based on JPEG-Compression-Derived Distortion2022

    • Author(s)
      K. Tsunomori, Y. Yamasaki, M. Kuribayashi, N. Funabiki, I. Echizen
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Survey on Vision Based Fake News Detection and Its Impact Analysis2022

    • Author(s)
      M. S. Raval, M. Roy, M. Kuribayashi
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Defense Against Adversarial Examples Using Beneficial Noise2022

    • Author(s)
      P. Raval, H. Khakhi, M. Kuribayashi, M. S. Raval
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Privacy Protection Against Automated Tracking System Using Adversarial Patch2022

    • Author(s)
      H. Takiwaki, M. Kuribayashi, N. Funabiki, M. S. Raval
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Classification of Screenshot Image Captured in Online Meeting System2022

    • Author(s)
      M. Kuribayashi, K. Kamakari, N. Funabiki
    • Organizer
      International IFIP Cross Domain Conference for Machine Learning & Knowledge Extraction (CD-MAKE2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 重み一定符号を用いたDNN電子透かしの検出法2022

    • Author(s)
      安井達哉, Malik Asad, 栗林稔
    • Organizer
      第21回情報科学技術フォーラム (FIT2022)
    • Related Report
      2022 Annual Research Report
  • [Presentation] 顔識別回避のためのパッチ型敵対的事例の生成及び堅牢性の検証2022

    • Author(s)
      瀧脇大登, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 5月
    • Related Report
      2022 Annual Research Report
  • [Presentation] 敵対的ノイズとJPEG圧縮由来の歪みの相関を用いた敵対的事例検出の研究2022

    • Author(s)
      角森健太, 山﨑裕真, 栗林稔, 舩曵信生, 越前功
    • Organizer
      信学技報, EMM 5月
    • Related Report
      2022 Annual Research Report
  • [Presentation] Fake News Detection and its impact analysis2022

    • Author(s)
      Mehul S. Raval, Mohendra Roy, Minoru Kuribayashi
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2022)
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] JPEG圧縮由来の歪み信号に対する応答特性に基づくAdversarial Examples検知手法2022

    • Author(s)
      角森健太, 山﨑裕真, 栗林稔, 舩曵信生, 越前功
    • Organizer
      2022年暗号と情報セキュリティシンポジウム (SCIS2022)
    • Related Report
      2021 Research-status Report
  • [Presentation] オンライン会議システム上の画面キャプチャによる情報漏洩の対策の一検討2022

    • Author(s)
      鎌苅康大, 栗林稔, 舩曵信生
    • Organizer
      2022年暗号と情報セキュリティシンポジウム (SCIS2022)
    • Related Report
      2021 Research-status Report
  • [Presentation] Detecting deepfake videos using digital watermarking2021

    • Author(s)
      A. Qureshi, D. Megias, M. Kuribayashi
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2021)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] A study of feature extraction suitable for double JPEG compression analysis based on statistical bias observation of DCT coefficients2021

    • Author(s)
      D. Takeshita, M. Kuribayashi, N. Funabiki
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2021)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] A study of feature extraction based on denoising auto encoder for classification of adversarial examples2021

    • Author(s)
      Y. Yamasaki, M. Kuribayashi, N. Funabiki, H. H. Nguyen, I. Echizen
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2021)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] A study of privacy protection of photos taken by wide-angle surveillance camera2021

    • Author(s)
      K. Nakai, M. Kuribayashi, N. Funabiki
    • Organizer
      Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2021)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] DISSIMILAR: Towards fake news detection using information hiding, signal processing and machine learning2021

    • Author(s)
      D. Megias, M. Kuribayashi, A. Rosales, and W. Mazurczyk
    • Organizer
      The 16th Int. Conf. Availability, Reliability and Security (ARES2021)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] White-box watermarking scheme for fully-connected layers in fine-tuning model2021

    • Author(s)
      M. Kuribayashi, T. Tanaka, S. Suzuki, T. Yasui, and N. Funabiki
    • Organizer
      9th ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec'21)
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research
  • [Presentation] 埋め込みロス関数なしのDNN電子透かしの収束に関する考察2021

    • Author(s)
      田中拓朗, 安井達哉, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 11月
    • Related Report
      2021 Research-status Report
  • [Presentation] 複数のAuto Encoderに対する応答特性を用いた敵対的事例の検出法2021

    • Author(s)
      山﨑裕真, 栗林稔, 舩曵信生, グエン フイ ホン, 越前功
    • Organizer
      信学技報, EMM 5月
    • Related Report
      2021 Research-status Report
  • [Presentation] DCT係数の統計的な偏り観測における二重JPEG圧縮履歴解析に適した特徴成分選出に関する考察2021

    • Author(s)
      竹下大地, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 5月
    • Related Report
      2021 Research-status Report
  • [Presentation] Adversarial Examples of Deep Learning System and Its Defense2021

    • Author(s)
      Minoru Kuribayashi
    • Organizer
      IEEE Computational Intelligence Society Chapter
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] 監視カメラにおける映り込み領域に対する顔検出とプライバシ保護に関する考察2021

    • Author(s)
      中井康貴, 栗林稔, 舩曵信生, 藤井爽平, 石原洋之
    • Organizer
      信学技報, EMM 3月
    • Related Report
      2020 Research-status Report
  • [Presentation] 複数のフィルタ強度によるCNN画像分類器の応答特性を用いた敵対的事例の検出法2021

    • Author(s)
      東亮憲, 栗林稔, 舩曵信生, Huy Hong Nguyen, 越前功
    • Organizer
      信学技報, EMM 3月
    • Related Report
      2020 Research-status Report
  • [Presentation] 顔領域に注目してファインチューニングした事前学習済みCNNモデルに基づくディープフェイク動画検出法2021

    • Author(s)
      河田健斗, 栗林稔, 舩曵信生
    • Organizer
      2021年暗号と情報セキュリティシンポジウム (SCIS2021)
    • Related Report
      2020 Research-status Report
  • [Presentation] DeepWatermark: embedding watermark into DNN model2020

    • Author(s)
      M. Kuribayashi, T. Tanaka, Nobuo Funabiki
    • Organizer
      Asia-Pacific Signal and Information Processing Association Anual Sumit and Conf. (APSIPA ASC 2020)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Classification of video recaptured from display device2020

    • Author(s)
      M. Kuribayashi, K. Kamakari, K. Kawata, N. Funabiki
    • Organizer
      Asia-Pacific Signal and Information Processing Association Anual Sumit and Conf. (APSIPA ASC 2020)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Detection of adversarial examples based on sensitivities to noise removal filter2020

    • Author(s)
      A. Higashi, M. Kuribayashi, N. Funabiki, H. Nguyen, I. Echizen
    • Organizer
      Asia-Pacific Signal and Information Processing Association Anual Sumit and Conf. (APSIPA ASC 2020)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] ブロックサイズの違いがDCT係数の頻度分布を利用したJPEG画像圧縮履歴解析に与える影響2020

    • Author(s)
      中嶋直也, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 11月
    • Related Report
      2020 Research-status Report
  • [Presentation] DM-QIMによるDNNモデルの全結合層に対する電子透かし法2020

    • Author(s)
      田中拓朗, 栗林稔, 舩曵信生
    • Organizer
      信学技報, EMM 9月
    • Related Report
      2020 Research-status Report
  • [Presentation] DM‐QIMによるディープニューラルネットワークの重みに対する電子透かし法2020

    • Author(s)
      田中拓朗, 安井達哉, 栗林稔, 舩曵信生
    • Organizer
      IEICE EMM研究会
    • Related Report
      2019 Research-status Report
  • [Presentation] DCT係数のヒストグラムを利用したJPEG画像のCNNに基づく編集履歴解析2020

    • Author(s)
      小浦啓太郎, 栗林稔, 舩曵信生
    • Organizer
      IEICE EMM研究会
    • Related Report
      2019 Research-status Report
  • [Presentation] ビデオカメラを用いたディスプレイ再撮により作成された違法動画の検出法2020

    • Author(s)
      鎌苅康大, 河田健斗, 栗林稔, 舩曵信生
    • Organizer
      IEICE EMM研究会
    • Related Report
      2019 Research-status Report
  • [Presentation] 非可逆圧縮の圧縮率の変化に伴う挙動解析に基づく敵対的事例の検知法2020

    • Author(s)
      東亮憲, 栗林稔, 舩曵信生, Huy Hong Nguyen, 越前功
    • Organizer
      IEICE EMM研究会
    • Related Report
      2019 Research-status Report
  • [Presentation] Improved DM-QIM Watermarking Scheme for PDF Document2019

    • Author(s)
      Minoru Kuribayashi and KokSheik Wong
    • Organizer
      International Workshop on Digital-forensics and Watermarking (IWDW2019)
    • Related Report
      2019 Research-status Report
    • Int'l Joint Research
  • [Presentation] DCT係数のヒストグラムの偏りを利用したJPEG画像の編集履歴解析2019

    • Author(s)
      小浦啓太郎, 栗林稔, 舩曵信生
    • Organizer
      コンピュータセキュリティシンポジウム (CSS2019)
    • Related Report
      2019 Research-status Report
  • [Presentation] JPEG圧縮による画像分類器の識別結果の変動解析に基づく敵対的事例の検知法2019

    • Author(s)
      東亮憲, 栗林稔, 舩曵信生, Nguyen Huy H., 越前功
    • Organizer
      コンピュータセキュリティシンポジウム (CSS2019)
    • Related Report
      2019 Research-status Report
  • [Presentation] 電子指紋符号の結託攻撃パラメータ推定のための特徴ベクトル導出及びその次元削減2019

    • Author(s)
      安井達哉, 栗林稔, 舩曵信生
    • Organizer
      コンピュータセキュリティシンポジウム (CSS2019)
    • Related Report
      2019 Research-status Report
  • [Book] Data Hiding and Its Applications: Digital Watermarking and Steganography2022

    • Author(s)
      D. Megias, W. Mazurczyk, M. Kuribayashi
    • Total Pages
      234
    • Publisher
      MDPI
    • ISBN
      9783036529363
    • Related Report
      2021 Research-status Report

URL: 

Published: 2019-07-04   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi