日浦 慎作

We evaluated the performance of our automated computerized scheme for determining the likelihood of malignancy of pulmonary nodules on high-resolution computed tomography (HRCT) and positron emission tomography (PET) scans. Our database contained 36 primary lung cancers and 9 benign nodules. After the nodule location was identified by a radiologist, the boundary surface of the nodule was segmented automatically using a spiral-scanning technique. Objective nodular features were assessed by quantitative analysis of the nodular shape and on gray-level histograms of the interior and exterior regions. The likelihood of malignancy was determined by a support vector machine. The performance of our technique in distinguishing between benign and malignant nodules was evaluated by receiver operating characteristic (ROC) analysis. The area under the ROC curve (AUC) value obtained by using HRCT features alone was 0.87. The integration of PET features into the AUC value resulted in a significant improvement from 0.87 to 0.94 (P< 0.01). The AUC value obtained from simultaneous selection of HRCT and PET features was 0.97. A statistically significant difference (P< 0.01) was observed between the result obtained by simultaneous HRCT and PET feature selection (AUC=0.97) and that by integration of PET features (AUC=0.94). Our automated computerized scheme for determining the likelihood of malignancy may help radiologists to differentiate between benign and malignant pulmonary nodules on HRCT and PET scans.

Calibration for calculating the external parameter of camera-projector system is important, to make high-precision 3-D measurement by 3-D endoscope using the grid active stereo. In this paper, we propose a new calibration method with a sphere calibration object in a geometrically flexible camera-projector system, and we examine the validity of the proposed method. As the result of comparing using conventional calibration with a plane calibration object, it is clarified that the proposed method resolves the scaling imprecision, that is a problem of the conventional method. Thus, the proposed method has higher potential of measurement accuracy. In addition, it is the advantage of the proposed method that precision estimation of calibration parameters with a low number of input images is possible.

The aim of this study is to improve the image quality using a post process rather than a correction process at acquisition time. We used a smoothing filter that is widely used on a compact digital camera. Especially for nuclear medicine, when we use a short acquisition time, we will get images that have a large increase in statistical noise. For those images, we validated the efficiency of the smoothing filter by assessing two characteristic parameters. In addition, we defined the best smoothing filter parameters to get stable images that reduced the influence of statistical noise.

本論文では撮像素子の各画素の受光感度分布をそれぞれランダムな形状に符号化することで,複数の画像を入力とする超解像処理を高性能化することができることを示す.既存の撮像素子の各画素の受光感度分布は方形であるため,画像の入力過程は原画像に対し方形を畳み込んだものをサンプリングしていることと等価である.このとき畳込みにより原画像の高空間周波数成分が失われることが超解像の性能を制限する要因となっている.そこで本研究では,撮像素子の上に細かい黒色粉末を散布することで,各画素の受光感度分布をランダムに符号化する.これに圧縮センシング分野で用いられているスパース正則化を組み合わせることで,少ない枚数の入力画像から高解像度な画像を復元し,既存の撮像素子を備えたカメラより超解像の性能が向上することを実機を用いた実験により示す.

ステレオ法はシーンに対して光を投影することなくシーンの奥行マップを得ることができる手法であるが,被写体上のテクスチャやエッジが繰返し模様であったりエピポーラ線と並行である場合に画像間の対応付けが不安定になるという問題がある.一方,錯乱円径が距離によって変化することを利用し距離を得るDepth from Defocusは,レンズの口径によって奥行精度が限定されるという欠点をもつ.そこで本研究では,ステレオカメラの左右のカメラに異なる合焦距離を与え,更に符号化開口を導入することでステレオ法とDepth from Defocusを融合し,両手法の利点を兼ね備えた安定度と精度をもつ奥行推定方法を提案する.

We have estimated scattering properties of participating media using rendered images. We have rendered multiple scattering in participating media using photon mapping. Then, we estimate the scattering properties by comparing the generated images and real images.

This paper describes a method to determine the direction of a light source and the distribution of diffuse reflectance from two images under different lighting conditions. While most inverse-rendering methods require 3 or more images, we investigate the use of only two images. Using the relationships between albedo and light direction at 6 or more points, we firstly show that it is possible to simultaneously estimate both of these if the shape of the target object is given. Then we extend our method to handle a specular object and shadow effect by applying a robust estimation method. Thorough experimentation shows that our method is feasible and stable not only for well controlled indoor scenes, but also for an outdoor environment illuminated by sunlight.

We propose a new non-contact measurement of cardiac beat from 3D shape information of body surface by using grid-based active stereo, and basically examine the validity of the proposed method. By simultaneous measurement with our proposed method and ECG, there are sufficient correspondences between peak intervals of inter-frame depth changes measured by our method and R-R intervals measured by ECG.

In this paper, we propose a method to accurately measure the shape of objects by suppressing indirect reflection such as interreflection or subsurface scattering. We use a modulation with M-sequence shifted along the line of the slit light to be accurately detected on the captured image in two ways. This method has two advantages; one is the characteristics of propagation of higher spatial frequency components and the other is geometric constraint between the projector and the camera. Prior to the measurement, epipolar constraint is obtained through calibration, and then the phase consistency is evaluated to suppress the interreflection. The value of cross-correlation is used to suppress the dilation of the light caused by the subsurface scattering.