Yasuhiro Ueda

High Resolution and High Dynamic Range Image Reconstruction from Differently Exposed Images

In this research, we propose a method to reconstruct image of high resolution and high dynamic range from small number of low resolution images with some saturated pixels. In this method the low resolution images are captured with different exposure times and at different positions and the high resolution image is created with reconstruction based super resolution processing. The position of the CCD sensor in the camera is controlled by a piezo actuator and at each capturing position the exposure time is also controlled in specific arrangement. And in the reconstruction method a high resolution image is recursively updated by comparing simulated low resolution images which are created from high resolution image with assumed optical degradation and its exposure time and the captured images. In this paper we first describe how to capture the low resolution images and how to create high resolution image from images which have some saturated pixels and next show the evaluation of the result image created from real captured images.

Image deblurring and denoising with non-local regularization constraint

In this paper, we investigate the use of the non-local means (NLM) denoising approach in the context of image deblurring and restoration. We propose a novel deblurring approach that utilizes a non-local regularization constraint. Our interest in the NLM principle is its potential to suppress noise while effectively preserving edges and texture detail. Our approach leads to an iterative cost function minimization algorithm, similar to common deblurring methods, but incorporating update terms due to the non-local regularization constraint. The dataadaptive noise suppression weights in the regularization term are updated and improved at each iteration, based on the partially denoised and deblurred result. We compare our proposed algorithm to conventional deblurring methods, including deblurring with total variation (TV) regularization. We also compare our algorithm to combinations of the NLM-based filter followed by conventional deblurring methods. Our initial experimental results demonstrate that the use of NLM-based filtering and regularization seems beneficial in the context of image deblurring, reducing the risk of over-smoothing or suppression of texture detail, while suppressing noise. Furthermore, the proposed deblurring algorithm with non-local regularization outperforms other methods, such as deblurring with TV regularization or separate NLM-based denoising followed by deblurring.

Super-Resolution Imaging for Automatic Visual Inspection Systems

The purpose of this research is to apply super-resolution processing to defect inspections in order to automate such inspections and increase their precision. Super-resolution processing, which is a method that creates a high-resolution image from multiple low-resolution images, could be an effective solution in environments where images of sufficient resolution cannot be obtained. However, reduction of noise in the generated images is necessary in order to conduct inspections reliably. Therefore, in this paper, we propose a noise reduction method for super-resolution processing, and then report on the defect inspection performance when this super-resolution processing is implemented.

Conceptual framework design for integrated visual inspection system

This paper describes a framework for automatic generation of an image processing algorithm that consists of preprocessing, feature extraction, classification and algorithm evaluation modules based on machine learning. With a view to applying the generated algorithm to industrial visual inspection system, we intend to offer a framework model equipped with the below-mentioned features. Also, we want to report on the experimental result of the offered model. 1.Automatically generate by machine learning an image processing algorithm to extract regions that have same characteristics as specified by users. 2.Generate in particular a high-precision image processing algorithm, improving the level of statistical separation between true and false defects that may cause a deterioration factor in classification accuracy. 3.Optimize an image improving filter sequence in preprocessing modules by means of GA (Genetic Algorithm).