An image compressing device includes: a histogram creating unit configured to create a histogram regarding an input image; an original type recognizing unit configured to recognize an original type of the input image using the histogram; and an image compressing unit configured to perform a compression process for creating a high-compressed image file from the input image and to control a compression mode of the compression process depending on at least the original type recognized by the original type recognizing unit.

A learning model generating device includes a first image reading device and a first control device. The first control device includes a processor and functions, through the processor executing a first control program, as a first segmenter, a learning model generator, and a first compressor. The first segmenter segments each of images of training prints obtained by reading performed by the first image reading device. The learning model generator learns segmented images to generate a first learning model for use in inferring a type of an image defect. The first compressor compresses each of the images of the training prints. The first segmenter segments each of compressed images obtained by compression performed by the first compressor. The learning model generator learns compressed and segmented images to generate a second learning model for use in inferring a type of an image defect.

Systems and methods for encoding high resolution data associated with a relatively large number of bits to an encoded form having a relatively reduced number of bits. The method includes, by a processor: receiving an input image comprising one or more high resolution objects. The method further includes, for each of the one or more high resolution objects: identifying an object family for that object and determining whether a reference table exists for the object family. If a reference table exists for the object family, the method includes determining a size of that object, and identifying a tag based on the size. The method also includes encoding that object to form an encoded object having a relatively reduced number of bits, associating the identified tag with the encoded object, and saving the encoded object.

In a color to grayscale image conversion particularly method suitable for processing color document images such as forms, the color image is analyzed to determined which of the red, green and blue channels are the most dominant, second most dominant, and least dominant channels, based on the amount of information contained in each channel. Then, coefficients are assigned to the three channels, where the coefficient for the most dominant channel is smaller than the coefficient for the second most dominant color channel, which is in turn smaller than the coefficient for the least dominant color channel. The grayscale pixel value is then calculated using a linear combination of the red, green and blue pixel values weighted by their assigned coefficients. In one example, the ratio of the coefficients for the least dominant, the second most dominant and the most dominant channels is 10:3:1.

A learning model generating device includes a first image reading device and a first control device. The first control device includes a processor and functions, through the processor executing a first control program, as a first segmenter, a learning model generator, and a first compressor. The first segmenter segments each of images of training prints obtained by reading performed by the first image reading device. The learning model generator learns segmented images to generate a first learning model for use in inferring a type of an image defect. The first compressor compresses each of the images of the training prints. The first segmenter segments each of compressed images obtained by compression performed by the first compressor. The learning model generator learns compressed and segmented images to generate a second learning model for use in inferring a type of an image defect.

An image forming apparatus includes: a hardware processor that: generates image data for composition; and divides the image data for composition into bands; an output memory; a storage that determines whether image data of each band coincides with image data of another band, secures a unique region, transfers the image data of the band to the unique region and associates the band with the unique region, secures common regions, transfers the image data of one of the bands to the common region and associates any one of the common regions to each of the bands; a reader that reads the image data from the region associated with each band and outputs the image data for composition; a composer that composes the image data for composition with the image data to be printed; and an image former that forms an image based on the composed image data.

An image processing method is provided for acquiring additional information from image information obtained by shooting a printed product on which the additional information is multiplexed by at least one of a plurality of different multiplexing methods, the method comprising: attempting decoding of the additional information from the image information by a plurality of different decoding methods corresponding to the plurality of different multiplexing methods; and outputting, by a unit, the additional information successfully decoded.

Systems and methods for encoding high resolution data associated with a relatively large number of bits to an encoded form having a relatively reduced number of bits. The method includes, by a processor: receiving an input image comprising one or more high resolution objects. The method further includes, for each of the one or more high resolution objects: identifying an object family for that object and determining whether a reference table exists for the object family. If a reference table exists for the object family, the method includes determining a size of that object, and identifying a tag based on the size. The method also includes encoding that object to form an encoded object having a relatively reduced number of bits, associating the identified tag with the encoded object, and saving the encoded object.

Image data representing a desktop image for a client device that is accessing the desktop remotely is compressed according to a method that preserves image fidelity in selected non-text regions. The method, which is carried out in a remote server, includes the steps of generating image data for the remote desktop image and analyzing different regions of the remote desktop image, identifying those regions of the remote desktop image that are text regions, selecting non-text regions of the remote desktop image for lossless compression based on a spatial relationship between the non-text regions and the text regions, compressing the image data using a lossless compression protocol for a portion of the image data corresponding to the selected non-text regions, and transmitting the compressed image data to the client device.

An image forming apparatus includes: a hardware processor that: generates image data for composition; and divides the image data for composition into bands; an output memory; a storage that determines whether image data of each band coincides with image data of another band, secures a unique region, transfers the image data of the band to the unique region and associates the band with the unique region, secures common regions, transfers the image data of one of the bands to the common region and associates any one of the common regions to each of the bands; a reader that reads the image data from the region associated with each band and outputs the image data for composition; a composer that composes the image data for composition with the image data to be printed; and an image former that forms an image based on the composed image data.