深度學習應用於印刷文件之來源鑑別

Abstract

在這個資訊與網路快速發達的時代，資訊取得非常容易並且透過數位內容的形式在網路上快速傳播著。雖然數位內容已被廣泛的應用，但由於印表機的容易取得及方便性，這些內容可以隨意的輸出成文件，印刷文件也常被有心人士利用，造成許多犯罪議題如：偽造文件、偽造貨幣、侵害著作權等。因此，如何運用有效率且適當的安全性測試方法及工具來鑑別印刷文件的來源是一個重大議題。

在相關的研究中，運用統計方法結合支持向量機開發鑑識系統針對文件中的文字及圖像分析後，鑑別文件的印表機來源，這樣的方法屬於機器學習中淺層學習的範疇，在特徵擷取、選擇等資料預處理的過程常是需要有人工參與。本研究基於深度學習能夠自動學習特徵的特性，運用卷積神經網路(Convolutional Neural Networks, CNN)來開發新的鑑識系統，利用卷積神經網路解決複雜的圖像分類問題。本研究期望可以透過卷積神經網路實現印表機來源鑑別，並且避免過多的人工參與及提升辨識率。透過兩系統鑑識結果的比較，探討深度學習是否能夠比淺層的機器學習有更好的表現。也透過實驗來探討印表機新舊程度、文字的不同樣本是否會對鑑識結果造成影響。

Due to the rapid development of the information technology and wide use of the Internet, Information is easily to be obtained in the form of digital format. Digital content can be freely printed into documents since the convenience and accessibility of the printers. On the other hand, printed documents are also illegally manipulated by some criminal issues such as: forged documents, counterfeit currency, copyright infringement, and so on. Therefore, how to develop an efficient and appropriate safety testing tool to identify the source of printed documents is an important task in the meantime.

Currently, the forensic system using the statistical methods and support vector machine technology has been able to identify the source printer for the text and the image documents. Such an approach belongs to the category of shallow machine learning with human interaction during the stages of feature extraction, feature selection and data pre-processing. In this study, a novel forensic system to solve the complex image classification problem is developed by Convolutional Neural Networks (CNNs) of deep learning which can learn the features automatically. We expect to implement CNNs for the printer source identification without too often human involvement with high accuracy. Through the experimental comparison of two above mentioned systems, the issues of whether the deep learning can do better than the shallow machine learning, whether the aging of printers can affect the identification, whether different region of the samples can affect the identification capability are all discussed.