本論文提出一個基於碎形編碼與區塊特徵的損壞的影像恢復方法。此機制可分為兩個步驟，一個是嵌入程序，另一個是偵測與恢復程序。在第一個程序裡，影像區塊的碎形碼與區塊特徵都會嵌入保護影像之像素值的最低有效位元（Least significant bit, LSB）中。在論文裡，我們分別採用兩種不同的區塊特徵來完成這個計畫。在第二個程序裡，可從驗證影像中取回驗證碼和兩個恢復資訊。藉由比較取回的驗證碼和從驗證影像計算出的驗證碼，可找出篡改區域。正確的恢復資訊可修補回近似原影像的像素值。由實驗結果可知，所提出的機制有良好的偵測與修補能力。

In this paper, an image tamper detection and recovery system based on block's characteristic and fractal coding is proposed. This scheme can be divided into two phases. One is the embedding procedure, and the other is the verifying and recovering procedure. In the first procedure, the fractal code and the characteristic of block both are embedded into the least significant bit (LSB) of the protection image pixels. In this paper, we respectively adopt two different characteristics of the block to achieve this approach. In the second procedure, the verification data and two recovering information can be extracted from an identified image. By using the comparison between the extracted verification code and the calculating verification code from the identified image, the tampered range can be found. The image pixels in the tampered range are able to be near recovered by using the correcting recovery information. The experiment results show that the proposed scheme can achieve the image detection and recovery.