Image processing apparatus, image recording apparatus, and image processing method

An image processing apparatus comprising:

• a histogram computing section which computes a histogram of pixel values of input image data;
• a fixed gradation conversion characteristic storing section which stores at least one fixed gradation conversion characteristic having a fixed characteristic and uniformly applicable to the input image data;
• a gradation conversion characteristic determining section which determines the fixed gradation conversion characteristic in a case where gradation conversion processing is applied to the input image data by deriving a cumulative histogram from the histogram computed by the histogram computing section, and by synthesizing the cumulative histogram derived and the fixed gradation conversion characteristic stored in the fixed gradation conversion characteristics storing section on the basis of information indicating a scene when the input image data has been picked up;
• a synthesis ratio storing section which stores a plurality of synthesis ratios to be used for the synthesis while associating with the information indicating the scene;
• and
• a gradation converting section which converts the gradation of the input image data on the basis of the gradation conversion characteristic determined by the gradation conversion characteristic determining section,
• wherein the gradation conversion characteristic determining section which determines a synthesis ratio to be used for the synthesis from the plurality of synthesis ratios stores in the synthesis ratio storing section on the basis of the information on the scene when the input image data has been picked up, and
• wherein the information indicating the scene includes at least one of information indicating a scene when the input image data has been picked up, flash information indicating whether or not a flash has been lit when the input image data has been picked up, information indicating whether or not the input image data has been picked up by automatic exposure, and photographing sensitivity information when the input image data has been picked up.

An image processing apparatus for applying a gradation conversion processing, the apparatus comprising:

• a histogram computing section which computes a histogram of pixel values of input image data;
• a memory storing a default gradation conversion table containing a standard characteristic, and an information on noise characteristic indicating what degree of noise is added, and in what form, when the input image data is picked up;
• a histogram correction section performing following steps:
• reading the default gradation conversion table stored in the memory;
• calculating a gradient of the default gradation conversion table by differentiating the default gradation conversion table;
• reading the information on noise characteristic which is a noise quantity, stored in the memory;
• estimating a noise quantity after the gradation conversion, which is a product of noise quantity and an amplification rate of noise after the gradation conversion, wherein the amplification rate of noise after gradation conversion is represented by the calculated gradient of the default gradation conversion table;
• calculating a reciprocal number of the noise quantity after the gradation conversion and determining the calculated reciprocal number as a frequency value limitation level to correct the histogram;
• and
• limiting a portion of the histogram exceeding the frequency value limitation level, thereby correcting the histogram;
• a histogram accumulating section which performs histogram accumulating process using the corrected histogram by the histogram accumulating section and obtaining an accumulated histogram;
• a gradation conversion table computing section computing a final gradation conversion table on the basis of the accumulated histogram obtained by the histogram accumulating section;
• and
• a gradation converting section performing the gradation conversion processing using the final gradation conversion table computed by the gradation conversion table computing section.

An image processing method for applying a gradation conversion processing, the method comprising:

• computing a histogram of pixel values of input image data by a histogram computing section;
• storing a default gradation conversion table containing a standard characteristic, information on noise characteristic indicating what degree of noise is added, and in what form, when the input image data is picked up in a memory;
• performing following steps for a histogram correction by a histogram correction section:
• reading the default gradation conversion table stored in the memory;
• calculating a gradient of the default gradation conversion table by differentiating the default gradation conversion table;
• reading the noise characteristic information which is a noise quantity, stored in the memory;
• estimating a noise quantity after the gradation conversion, which is a product of the noise quantity and an amplification rate of noise after the gradation conversion, wherein the amplification rate of noise after gradation conversion is represented by the calculated gradient of the default gradation conversion table;
• calculating a reciprocal number of the noise quantity after the gradation conversion and determining the calculated reciprocal number as a frequency value limitation level to correct the histogram;
• and
• limiting a portion of the histogram exceeding the frequency value limitation level, thereby correcting the histogram;
• performing a histogram accumulating process by inputting the corrected histogram to a histogram accumulating section, thereby obtaining an accumulated histogram;
• computing a final gradation conversion table from the accumulated histogram obtained by the histogram accumulating section by storing a gradation conversion table computing section;
• and
• performing a gradation conversion processing using the computed final gradation conversion table by a gradation converting section.

An image processing method for applying a gradation conversion processing, the method comprising:

• performing following steps for an image processing circuit;
• computing a histogram of pixel values of input image data by a histogram computing section;
• storing a plurality of fixed gradation conversion characteristics each having a fixed characteristic and applicable to the input image data and a plurality of gradation synthesis ratios while associating with the information indicating the scene when the input image data is picked up;
• deriving an accumulated histogram from the computed histogram by a histogram accumulating section;
• reading the plurality of fixed gradation conversion characteristics and the plurality of gradation synthesis ratios in accordance with the information indicating the scene;
• determining the gradation conversion characteristic in a case where gradation conversion processing is applied to the input image data by synthesizing the cumulative histogram derived and the fixed gradation conversion characteristic read from the memory on the basis of a gradation synthesis ratio read from the memory;
• and
• converting the gradation of the input image data on the basis of the determined gradation conversion characteristic,
• wherein the information indicating the scene includes at least one of information indicating a scene when the input image data has been picked up, flash information indicating whether or not a flash has been lit when the input image data has been picked up, information indicating whether or not the input image data has been picked up by automatic exposure, and photographing sensitivity information when the input image data has been picked up.