摘要:
An apparatus and a method as well as a storage device of the present invention transform intensity levels with consideration of both features of the intensity levels in an input image and human visual characteristics, wherein the present invention includes the steps of inputting wider dynamic range digital image data, generating a histogram of intensity levels of the image data, generating a histogram equalization LUT by cumulating the histogram, generating a visual characteristics LUT with reference to the histogram equalization LUT, generating a combined LUT by combining the histogram equalization LUT and the visual characteristics LUT, and correcting the image data.
摘要:
An image processing apparatus and method and a recording medium therefor control the occurrence of a change in a hue. The histogram of a luminance signal Y is generated in a first step S1. In a second step S2, the histograms of the luminance signals Y are accumulated to generate a cumulative histogram, and the cumulative histogram is subjected to logarithmic approximation thereby to generate a lookup table (LUT) for correcting luminance signals. In a third step S3, a color-difference signal correction parameter LUT is generated by referring to the LUT for correcting luminance signals. In a fourth step S4, an input luminance signal Y is applied to the LUT for correcting luminance signals thereby to acquire a corrected luminance signal Y0. In a fifth step S5, the input luminance signal Y is applied to a color-difference signal correction parameter LUT so as to acquire a correction parameter k. In a sixth step S6, input color-difference signals Cr and Cb are multiplied by the correction parameter k to acquire corrected color-difference signals Cr0 and Cb0.
摘要:
An apparatus and a method as well as a storage device of the present invention transform intensity levels with consideration of both features of the intensity levels in an input image and human visual characteristics, wherein the present invention includes the steps of inputting wider dynamic range digital image data, generating a histogram of intensity levels of the image data, generating a histogram equalization LUT by cumulating the histogram, generating a visual characteristics LUT with reference to the histogram equalization LUT, generating a combined LUT by combining the histogram equalization LUT and the visual characteristics LUT, and correcting the image data.
摘要:
An apparatus and a method as well as a storage device of the present invention transform intensity levels with consideration of both features of the intensity levels in an input image and human visual characteristics, wherein the present invention includes the steps of inputting wider dynamic range digital image data, generating a histogram of intensity levels of the image data, generating a histogram equalization LUT by cumulating the histogram, generating a visual characteristics LUT with reference to the histogram equalization LUT, generating a combined LUT by combining the histogram equalization LUT and the visual characteristics LUT, and correcting the image data.
摘要:
The invention relates to an image data expansion processor for digital still camera. Two blocks of 8-bit luminance data and one block of 8-bit color difference data are repeatedly output from a compressing/expanding unit in the above order and are supplied to a buffer memory. Two blocks of 16-bit data to be written are generated corresponding to the two blocks of 8-bit luminance data and one block of 8-bit color difference data, are sequentially supplied from the buffer memory to a frame memory in units of block and are written to the frame memory. Each of luminance data and color difference data is not written to the frame memory in units of block but 16-bit data to be written composed of these luminance data and color difference data is written to the frame memory in units of block. The number of blocks written to the frame memory can be reduced and total time required for switching an address in the direction of lines can be reduced.
摘要:
This invention relates to an image data processing method or the like capable of obtaining a preferable reproduction image based on processed image data even if the image data is short of the unit of block on reading the image data from the memory means in the unit of block. When image data of “M lines×N pixels” written in a memory is read in a unit of block consisting of the predetermined number of pixels and the read image data is subjected to a compression coding process or the like, if data of two lines and data of four pixels per line are short, data of two pixels per line is added to the left end side of the image by using data (1,1) (2,1) . . . (M,1) in the left end side thereof and data of two pixels is added to the right end side of the image by using data (1,N) (2,N) . . . (M,N) in the right end side thereof. Data of one line is added to the upper end side of the image by using data (1,1) (1,2) . . . (1,N) in the upper end side thereof and data of one line is added to the lower end side of the image by using data (M,1) (M,2) . . . (M,N) in the lower end side thereof. Thus, an image data added part is inconspicuous, and deterioration in picture quality caused by a compressing and reduction in the data compression ratio can be prevented.
摘要:
The present invention is directed to a digital still camera (1) using CCD (Charge Coupled Device) image sensor as image pick-up device. This camera (1) comprises an image memory (18) in which plural images obtained by picking up images of the same object by different exposure values are stored, an image data adding circuit (17) for synthesizing the plural images stored in the image memory (18) to generate synthetic image in which dynamic range is broad, a Y gradation correction circuit (24) and a multiplier circuit (26) for compressing gradation of the synthetic image generated by the image data adding circuit (17), and a controller (29) for generating gradation conversion table for carrying out gradation compression. The controller (29) generates gradation conversion table on the basis of synthetic curve in which line determined on the basis of ratio between the number of gradations of synthetic image and the number of gradations of output image and histogram indicating luminance distribution of synthetic image are synthesized.
摘要:
This invention relates to an image data processing method or the like capable of obtaining a preferable reproduction image based on processed image data even if the image data is short of the unit of block on reading the image data from the memory means in the unit of block. When image data of “M lines×N pixels” written in a memory is read in a unit of block consisting of the predetermined number of pixels and the read image data is subjected to a compression coding process or the like, if data of two lines and data of four pixels per line are short, data of two pixels per line is added to the left end side of the image by using data (1,1) (2,1) . . . (M,1) in the left end side thereof and data of two pixels is added to the right end side of the image by using data (1,N) (2,N) . . . (M,N) in the right end side thereof. Data of one line is added to the upper end side of the image by using data (1,1) (1,2) . . . (1,N) in the upper end side thereof and data of one line is added to the lower end side of the image by using data (M,1) (M,2) . . . (M,N) in the lower end side thereof. Thus, an image data added part is inconspicuous, and deterioration in picture quality caused by a compressing and reduction in the data compression ratio can be prevented.
摘要:
This invention relates to an image data processing method or the like capable of obtaining a preferable reproduction image based on processed image data even if the image data is short of the unit of block on reading the image data from the memory means in the unit of block. When image data of “M lines×N pixels” written in a memory is read in a unit of block consisting of the predetermined number of pixels and the read image data is subjected to a compression coding process or the like, if data of two lines and data of four pixels per line are short, data of two pixels per line is added to the left end side of the image by using data (1,1) (2,1) . . . (M,1) in the left end side thereof and data of two pixels is added to the right end side of the image by using data (1,N) (2,N) . . . (M,N) in the right end side thereof. Data of one line is added to the upper end side of the image by using data (1,1) (1,2) . . . (1,N) in the upper end side thereof and data of one line is added to the lower end side of the image by using data (M,1) (M,2) . . . (M,N) in the lower end side thereof. Thus, an image data added part is inconspicuous, and deterioration in picture quality caused by a compressing and reduction in the data compression ratio can be prevented.