Imaging device, image processing method, image processing program and semiconductor integrated circuit

Abstract

A line-by-line filter process is performed to extract luminance data and color carrier data from an original image (i.e., a CCDRAW image) obtained from a single-sensor color imaging element, without separating the original image into R, G, and B data corresponding to a Bayer array. A color difference data demodulation process and a median noise removal process are performed with respect to the color carrier data to obtain color difference data. Band limitation filter processes having different characteristics are performed with respect to the luminance data and the color difference data, respectively. Moreover, separate resizing processes are performed with respect to the luminance data and the color difference data, respectively. A modulation process is performed to convert the resized color difference data to color carrier data. A color array reproduction process is performed to recombine the color carrier data and the resized luminance signal.

Description

technical field [0001] The present invention relates to an imaging device, an image processing method, an image processing program, and a semiconductor integrated circuit equipped with an image processing function, and in particular to an image suitable for reduction processing, enlargement processing, and deformation correction of an original image obtained from a single-board color imaging element Data size conversion processing device, electronic still picture camera, image data size conversion processing program, semiconductor integrated circuit, etc. Background technique [0002] There is known a digital camera that captures an image of a subject passing through a photographic lens with an imaging element such as a CCD or a MOS sensor, and then performs predetermined image processing on image data output from the imaging element. A general digital camera converts the image signal obtained from the imaging element into 10-bit, 12-bit, or 14-bit digital data, performs var...

AI Technical Summary

Problems solved by technology

[0013] In the technologies of patent documents 1 to 3, when resizing the RAW data of RGB as the original image, color separation into independent R, G, and B data in the initial processing is adopted, or by mixing pixels of the same color The method of interval extraction, so when color separation is performed on the spatially configured RGB data in the RAW data, due to the interval extraction realized by sampling and pixel addition, it turns back to the high frequency near the low frequency Nyquist frequency Components cannot be removed by LPF, so high-frequency artifacts will be generated on the reduced RAW data that is reconfigured after reduction

As a result, false colors appear in the image, deterioration of the sense of resolution occurs, and there is a disadvantage that a reduced RAW image that fully utilizes the imaging characteristics of the original image cannot be obtained.

[0014] In addition, in the technology of Patent Document 4, the reduction and resizing process is performed after the image processing of the actual size, but the reduction and resizing process is a process of reducing the number of recording pixels, and the image processing is performed by thinning by the reduction and resizing process. The lost data becomes redundant processing, from the point of view of power saving, the efficiency is poor

Also, when it is installed in hardware such as LSI, since it is processed via an external memory buffer, the power consumption of the I / O buffer of the LSI is in proportion to the amount of processing.

Images