A method for generating an image processing filter includes: adjusting; and extracting. The adjusting inputs first training image data into a neural network to generate output image data, calculates an evaluation value based on a loss function using the output image data and second training image data, and adjusts a convolution filter so as to reduce the evaluation value. The extracting extracts data from the adjusted convolution filter as data for the image processing filter. A first training image includes noise and reproduces a test pattern. A second training image includes reduced noise and reproduces the test pattern. The loss function includes a first term and a second term. The first term specifies a magnitude of a difference between the output image data and the second training image data. The second term grows smaller as symmetry of the convolution filter relative to a filter axis of symmetry increases.

A method for generating an image processing filter includes: adjusting; and extracting. The adjusting inputs first training image data into a neural network to generate output image data, calculates an evaluation value based on a loss function using the output image data and second training image data, and adjusts a convolution filter so as to reduce the evaluation value. The extracting extracts data from the adjusted convolution filter as data for the image processing filter. A first training image includes noise and reproduces a test pattern. A second training image includes reduced noise and reproduces the test pattern. The loss function includes a first term and a second term. The first term specifies a magnitude of a difference between the output image data and the second training image data. The second term grows smaller as symmetry of the convolution filter relative to a filter axis of symmetry increases.

In some examples, luminance-biased sharpening for thermal media printing may include converting an input image to a grayscale luminance representation. For each pixel of a plurality of specified pixels of the converted input image, a sharpening lightness value may be determined. Further, a ratio of the sharpening lightness value to a corresponding original lightness value may be determined. A resulting sharpened pixel may be determined by applying a corresponding value of the determined ratio to each of the specified pixels. A dark correction factor may be applied to the resulting sharpened pixels that are darkened and a light correction factor may be applied to the resulting sharpened pixels that are lightened. Based on application of the dark correction factor and the light correction factor, a sharpened output image corresponding to the input image may be generated.

Disclosed is a signal processing device and an image display apparatus including the same. In the signal processing device and the image display apparatus according to the present disclosure, a High Dynamic Range (HDR) processor receives an image signal and adjust a luminance of the image signal, and a reduction unit configured to amplify the adjusted luminance of the image signal and increase a resolution of the grayscale of the image signal to generate an enhanced image signal, wherein the enhanced image signal provides an increased luminance and grayscale resolution of the image signal while maintaining high dynamic range within the displayed HDR image. Accordingly, expression of high grayscale of a received image may improve.

Disclosed is a signal processing device and an image display apparatus including the same. The signal processing device and the image display apparatus comprise: a first reduction unit to receive a image signal and reduce noise of the received image signal, and a second reduction unit to perform grayscale amplification based on the image signal from the first reduction unit, wherein the second reduction unit is configured to perform the grayscale amplification so that upper-limit level of grayscale of the image signal from the first reduction unit is greater than upper-limit level of grayscale of an OSD signal. Accordingly, OSD area may be uniformly displayed regardless of ambient luminance.

Techniques for reducing or eliminating image banding in an ink-jet image are provided. In an example, a method of operating a printer to reduce or eliminate image banding can include generating command profile for printing a given image, applying a filter to the command profile to provide a filtered profile, and dispensing ink from a printhead of the printer based on the filtered profile. In certain examples, the filter can randomize droplet sizes of ink dispensed while executing the printing to reduce or eliminate image banding.

Techniques for reducing or eliminating image banding in an ink-jet image are provided. In an example, a method of operating a printer to reduce or eliminate image banding can include generating command profile for printing a given image, applying a filter to the command profile to provide a filtered profile, and dispensing ink from a printhead of the printer based on the filtered profile. In certain examples, the filter can randomize droplet sizes of ink dispensed while executing the printing to reduce or eliminate image banding.

Novel tools and techniques are provided for implementing identification and removal of noise from documents, and, more particularly, to methods, systems, and apparatuses for implementing identification and removal of noise from financial documents using one or more machine learning algorithms. In various embodiments, computing system might receive a document. The computing system might detect, using one or more machine learning algorithms, that noise exists in the document. Based on the detection that noise exists in the document, the computing system might remove the noise from the document. Once the noise is removed from the document, the computing system might generate a copy of the document with the noise removed while retaining important or useful information contained in the document.

Novel tools and techniques are provided for implementing identification and removal of noise from documents, and, more particularly, to methods, systems, and apparatuses for implementing identification and removal of noise from financial documents using one or more machine learning algorithms. In various embodiments, computing system might receive a document. The computing system might detect, using one or more machine learning algorithms, that noise exists in the document. Based on the detection that noise exists in the document, the computing system might remove the noise from the document. Once the noise is removed from the document, the computing system might generate a copy of the document with the noise removed while retaining important or useful information contained in the document.

An image processing apparatus includes a processor configured to: acquire image information obtained by an image reading apparatus including an illumination unit having an uneven balance in an amount of light in a sub-scanning direction; and specify a position of a folding line in a sheet from the acquired image information. The acquiring by the processor includes acquiring first image information obtained by reading the sheet in a first state in which the folding line in the sheet is positioned in a direction intersecting the sub-scanning direction of the image reading apparatus, and second image information obtained by reading the sheet in a second state in which the sheet is directed to an opposite side to a side in the first state. The specifying by the processor includes specifying the position of the folding line in the sheet based on the first image information and the second image information.