Kiyoaki Atsuta

A New Content-based Image Retrieval Using Color Correlogram and Inner Product Metric

The color correlogram is a simple statistical descriptor of a color image that has been widely used for content-based image retrieval (CBIR) systems. To measure similarity between two images using the correlogram, the traditional approaches use the relative distance. In this paper, to improve performance of the CBIR systems, the inner product metric is used to measure similarity of images instead of the relative distance. Results of experiments proved that the CBIR using the inner product metric has better performance than the one using the relative distance.

A New Similarity Measure for Content-Based Image Retrieval Using the Multidimensional Generalization of the Wald-Wolfowitz Runs Test

This paper proposes a new similarity measure for the content-based image retrieval (CBIR) systems. The similarity measure is based on the multidimensional generalization of the Wald-Wolfowitz (MWW) runs test and the k-means clustering algorithm. The performance comparisons between the proposed method and the current CBIR method based on MWW runs test were performed, and it can be seen that the proposed methods outperform the current method in the sense that the proposed method provides higher performance than the current method for the same computational time.

An Improved Wavelet-based Watermarking Method Using the Mathematical Morphology

In this paper, we propose an improved wavelet- based watermarking method using the mathematical morphology. The wavelet coefficients of host image are firstly selected as the significant coefficients by a statistical predetermined threshold. The mathematical operations are then performed to tame some small perturbations on the selected coefficients caused by the manipulations or attacks. Moreover, the pixel-wise masking model which exploits the characteristics of human visual system is employed for improving watermark invisibility and robustness. The experimental results verify that our method is more robust and effective than the existing method.

A theory of image restoration for linear spatial degradation using multiresolution analysis

In this theory it is assumed that an operator representing linear degradation is unknown but a pair of original images and its degraded image is given. The two images are projected onto several image spaces whose spatial frequencies are included in specified bands using the MRA. By comparison between the two projected images in the image spaces, demodulation coefficients are obtained which are used to restore the degraded image to the original one. In the process of obtaining the coefficients a new image distance is used which is a comparable measure that would correspond to the way the human visual system assesses. The obtained demodulation coefficients can be used to restore other degraded images. Examples of image restorations support the effectiveness of the proposed method.

An adaptive digital watermark based on a tree structure using the human visual system

Digital watermarking has been proposed as a way to claim ownership. In this paper, we propose a blind watermarking that does not require an original image in the detection process. We construct an image-dependent watermark and embed it in the most significant discrete wavelet transform (DWT) coefficients, which are determined by using hierarchical tree structure. When the locations of such coefficients are changed due to image manipulations, the tree structure allows the watermark detector to recover the set of embedded coefficients. We also propose a new frequency band-wise weight of the human visual system (HVS). With this weight, we can increase the robustness of watermarks without deterioration of perceptibility. In our experiments, the robustness of the proposed scheme is better than the one without weighting.

Nonlinear digital filters for image restoration using Volterra series

In this paper, we propose new nonlinear digital filters for image processing which are represented by discrete Volterra series. Firstly a design theory of nonlinear filters is formulated as an identification problem of unknown filters using pairs of input and output images of the filters. Secondly a nonlinear filter for image restoration is designed as an example of the proposed design method. Finally the designed filter is used to restore another image degraded in the same manner. It supports that the proposed design method is quite efficient.

A model of human estimation of 3D shape from shading information using zero crossings

Proposes a model of human estimation of 3D shape from a single image using estimation theory. Firstly it is seen, using an example, that most human subjects estimate a 3D shape from the single image which is physically contradictory. Secondly it is seen that in order to explain the human false estimate it is necessary that in human estimation a given image is divided into several regions and different illuminant directions are used depending on the regions. Thirdly a model of human estimation of 3D shape and an algorithm of 3D shape estimation following the model are proposed.<<ETX>>

Estimation of 3‐D shape from shading by using a constraint at zero‐crossing

This paper describes estimations of the shape of a three-dimensional (3-D) object from a gray-level 2-D image, i.e., “shape from shading.” The fact that the Ikeuchi-Horn iterative method produces an error when this is applied to the estimation of an object having a concave surface is shown. A new method using a constraint on a zero-crossing is proposed which can be applied to such an object. Examples of application of the proposed method are shown to confirm its effectiveness, from a case where several interpretations are possible from a single gray-level 2-D image, the proposed method can estimate all of them. Examples of such cases also are shown.

Restoration theory for digital images under additive noises by a constrained generalized inverse matrix

The image restoration by using Moore-Penroses generalized inverse (MPGI) is quite simple to implement, but it has vital defects: (a) it must compute the eigenvalues and eigenvectors of a large scale matrix, (b) restored images by this method are badly affected by the additive noise. In this paper, a new image restoration by the MPGI constrained by a quality measure of restored images is proposed, which can remove the defects of the current MPGI-restoration. Prior information the proposed image restoration requires is only a matrix which represents the point-spread function of degradation.