Kurato Maeno

New semi-fragile image authentication watermarking techniques using random bias and nonuniform quantization

Semi-fragile watermarking techniques aim at detecting malicious manipulations on an image, while allowing acceptable manipulations such as lossy compression. Although both of these manipulations are considered to be pixel value changes, semi-fragile watermarks should be sensitive to malicious manipulations but robust to the degradation introduced by lossy compression and other defined acceptable manipulations. In this paper, after studying the characteristics of both natural images and malicious manipulations, we propose two new semi-fragile authentication techniques robust against lossy compression, using random bias and nonuniform quantization, to improve the performance of the methods proposed by Lin and Chang.

A new semi-fragile image authentication framework combining ECC and PKI infrastructures

Presents a new semi-fragile framework aiming at extending public key signature scheme from message level to content level. The content signing procedure includes signature generation and watermark embedding while the content authentication procedure includes watermark extraction and signature verification. One main unique contribution is the novel use of error correction coding (ECC) to address the incidental distortions caused by some acceptable manipulations such as lossy compression. Another unique feature is integration of PKI security infrastructure and the hashing mechanism to achieve security and short signatures/watermarks. In the signing procedure, block-based invariant features are extracted from the image content and then encoded by ECC to obtain their corresponding parity check bits (PCB). All PCBs are then embedded back into image as watermarks for the purpose of authentication and locating content alteration. In addition, codewords are concatenated, hashed and finally signed by content owners private key to generate a global cryptographic signature. The authentication procedure is the inverse procedure of signing except using content owners public key for signature verification.

Non-contact heart rate detection using periodic variation in Doppler frequency

Remote heart rate detection without body-attached probes is a promising technology for health care, monitoring of elderly people, emergency, and security. In this paper, we use a continuous wave (CW) microwave Doppler radar. It is important to eliminate the effect of body movement that is irrelevant to heartbeat such as respiration. In general, the displacements of them are larger than those of heartbeat. Therefore, we focus on the periodic variation of velocity of body movement due to heartbeat rather than the displacement variation of it. We detect a heart rate from a part of the wavelet frequency components with high periodicity. As a result of performance evaluation, our system enables to extract more accurate heartbeat interval than the traditional approach using the periodicity of an original Doppler signal.

New semifragile image authentication watermarking techniques using random bias and nonuniform quantization

Semi-fragile watermarking techniques aim at detecting malicious manipulations on an image, while allowing acceptable manipulations such as lossy compression. Although both of these manipulations are considered to be pixel value changes, semi-fragile watermarks should be sensitive to malicious manipulations but robust to the degradation introduced by lossy compression and other defined acceptable manipulations. In this paper, after studying the characteristics of both natural images and malicious manipulations, we propose two new semi-fragile authentication techniques robust against lossy compression, using random bias and nonuniform quantization, to improve the performance of the methods proposed by Lin and Chang.

Human detection algorithm for Doppler radar using prediction error in autoregressive model

Using a Doppler radar as a motion sensor is promising for monitoring human daily activities. Various detection algorithms for human activities and vital signs have been proposed; however, most require a controlled environment and do not allow for the presence of moving objects in the sensing area. In actuality, motor-actuated appliances such as electric fans and heaters are often noise sources in homes. The velocities of such appliances overlap with those of a person and can cause a Doppler effect even without human presence. We resolve this issue in the current study by analyzing motion regularity. To detect the irregular motions that characterize a persons presence, we utilize the prediction errors from an autoregressive model. A performance evaluation of the proposed algorithm shows that high accuracy is achieved for human detection under disturbances caused by electric appliance motions. Moreover, our method outperforms the conventional frequency-based method that uses power spectra of the Doppler signal.

Statistical Recognition of Aspiration Presence

A study on the healthcare application is very important for the solitary death in aging society. Many previous works had been proposed a detection method of aspiration using the non-contact radar. But the works are only in subjects with sitting in a chair. We consider that user falls down in the state when he happen abnormal situation as daily life. In this study, we focus on the detection of “aspiration” or “apnea” for the lying position, because the final decision of the life or death is aspiration. As initial stage of the system, we propose the recognition method for the presence of aspiration with lying position under the low-disturbance environment from microwave Doppler signals by using support vector machine (SVM).

Dynamic respiratory modeling for non-contact live monitoring by particle filter approach

Although a Doppler radar detects the respiratory motion in the non-contact way, there is a problem of the performance decrement in proportion to the distance between a radar and a human body due to the reduction of the reflected rate. A model-based method is proposed to evaluate the respiratory presence in such cases. This model expresses the chest-wall displacement. The expression is composed of the periodic function which has five parameters. This model has a new property that the inspiration and the expiration are given by each of the two independent parameters. Therefore the problem is solved by tracking the radar outputs using the particle filter framework. The experiment was carried out to the six subjects at the distance 3.25m. The result showed that there was the statistically significant difference between the evaluation value of the unattended state and that of the attended one. In addition, the estimated model was favorably compared with the reference data which was measured by the high-precision displacement sensor. Consequently, the efficacy of the proposed method for the long distance (>3.00m) and that of the proposed respiratory model are established.