西华大学学报（自然科学版）2026，Vol.45Issue(3)：83-92,10.DOI:10.12198/j.issn.1673-159X.5383
基于特征谱图融合的音频信号复制粘贴篡改检测算法
Copy-paste Tamper Detection Algorithm for Audio Signals Based on Feature Spectrogram Fusion
摘要
Abstract
Copy-and-paste tampering is characterized by its manipulability and imperceptibility.Mean-while,the existence of various post-processing operations makes audio tampering detection more difficult.For audio copy-paste tampering with post-processing,the paper proposes an audio copy-paste tampering de-tection algorithm based on the fusion of feature spectrograms.Firstly,the audio signal is converted into Hil-bert spectrogram and Mel spectrogram respectively,and the two feature spectrogram images are weighted and fused to obtain a more robust feature spectrogram image.Further the key points are extracted from the fused feature image by utilizing scale invariant feature transform and matching the keypoints.Then the du-plicate regions on the image are identified with an agglomerative hierarchical clustering algorithm to imple-ment audio copy-paste tampering detection.The experimental results show that the proposed method is highly robust against the various common post-processing attacks.Compared with previous state-of-the-art audio signal copy-paste tampering detection methods,the proposed method in this paper has higher detec-tion precision,recall and robustness.关键词
复制粘贴篡改/特征谱图融合/尺度不变特征变换/凝聚层次聚类/伪造定位Key words
copy-paste tampering/feature spectrogram fusion/scale invariant feature transform/agglomerative hierarchical clustering/forged location分类
信息技术与安全科学引用本文复制引用
孙先峰,邱新,彭锴,王冬宇,史沧红,牛宪华,齐超..基于特征谱图融合的音频信号复制粘贴篡改检测算法[J].西华大学学报（自然科学版）,2026,45(3):83-92,10.基金项目
四川省自然科学基金项目(2023NSFSC0470) （2023NSFSC0470）
国家自然科学基金项目(62171387). （62171387）
