Abstract: The present subject matter describes a method to detect anomaly in an environment based on AI techniques. The method comprises receiving one or more data representations of one or more objects present in an environment. A first-type of information is captured from a first-area within the one or more data representations. A second-type of information from a second-area different than the first area in the data representations is also captured. A third information is generated from the first information, said third information corresponding to predicted information for the second area using one or more artificial-intelligence models for evaluating the second information. The third information is compared with the second information to determine abnormality with respect to state or operation of one or more objects within the environment.

Abstract: The present subject matter describes a method to detect anomaly in an environment based on AI techniques. The method comprises receiving one or more data representations of one or more objects present in an environment. A first-type of information is captured from a first-area within the one or more data representations. A second-type of information from a second-area different than the first area in the data representations is also captured. A third information is generated from the first information, said third information corresponding to predicted information for the second area using one or more artificial-intelligence models for evaluating the second information. The third information is compared with the second information to determine abnormality with respect to state or operation of one or more objects within the environment.

Abstract: An image processing apparatus includes a scanner that acquires an image, and a processor that changes areas of a plurality of dot patterns in accordance with gradation values in the image using a plurality of dither matrices, and generates a binary image based on the plurality of dot patterns. A change rate of each of the areas differs from each other. The plurality of dither matrices include a first dither matrix and a second dither matrix. A first change rate of a first area of a plurality of first dot patterns formed based on the first dither matrix and a second change rate of a second area of a plurality of second dot patterns formed based on the second dither matrix differ.

Abstract: An image processing apparatus includes a scanner that acquires an image, and a processor that changes areas of a plurality of dot patterns in accordance with gradation values in the image using a plurality of dither matrices, and generates a binary image based on the plurality of dot patterns. A change rate of each of the areas differs from each other. The plurality of dither matrices include a first dither matrix and a second dither matrix. A first change rate of a first area of a plurality of first dot patterns formed based on the first dither matrix and a second change rate of a second area of a plurality of second dot patterns formed based on the second dither matrix differ.

Abstract: An image processing apparatus capable of improving image quality of binarized image data is provided. The image processing apparatus includes a generation part that generates a binarized image data. The binarized image data comprises a first cell, containing a first dot pattern in which an outside pixel density is higher than an inside pixel density; and a second cell, containing a second dot pattern in which an inside pixel density is higher than an outside pixel density.

Abstract: The invention provides an image forming device, which can generate a multi-valued image based on a binary image. An image forming device includes: an image acquisition part for acquiring a binary image; and a multi-valued image generation part generating a multi-valued image including a specific pixel based on pixel values of pixels within a first image region with a first size including the specific pixel and pixel values of pixels within a second image region with a second size different from the first size including the specific pixel in the binary image.

Abstract: A photograph image-processing method comprises the steps of entering a film image for reading a film image by an image pickup device and generating color image data, generating concentration histograms for each of RGB color components from the color image data, relative-stretch-processing the concentration histograms, calculating and deriving a relative stretch ratio which maximizes the RGB superimposed area which has undergone relative-stretch processing, calculating and deriving an evaluation value of reliability with respect to the relative stretch ratio, calculating and deriving a correction stretch ratio based on the evaluation value, and conversion-processing RGB components of each pixel of the color image data in accordance with the correction stretch ratio.

Abstract: A method of detecting the base concentration of a film comprises the steps of generating a concentration histogram from the film image in each of the RGB color components, shifting and stretching the histograms in the other colors towards the histogram in the reference color in the concentration axial direction, calculating an superimposed area between the two histograms in the two color components after the stretching process, and finding the shift rate and the stretch ratio when a total of the respective superimposed areas calculated is the greatest value. As the minimum of the lowest concentration in the histograms at the shift rate and the stretch ratio determined is specified to represent the location of the base concentration in the original histogram, the base concentration in each color component of the film can be calculated from the location of the base concentration in the histogram.

Abstract: A method of detecting the base concentration of a film is provided comprising the steps of generating a concentration histogram from the film image in each of the RGB color components read out with an image pick up device, shifting and stretching the histograms in the other colors towards the histogram in the reference color in the concentration axial direction, calculating an superimposed area between the two histograms in the two color components after the stretching process, and finding the shift rate and the stretch ratio when a total of the respective superimposed areas calculated is the greatest value. As the minimum of the lowest concentration in the histograms at the shift rate and the stretch ratio determined is specified to represent the location of the base concentration in the original histogram, the base concentration in each color component of the film can be calculated from the location of the base concentration in the histogram.

Abstract: From color image data read by an image pickup device, concentration histograms are generated for each of RGB color components, and with the specific color concentration histogram used as reference, other color concentration histograms are shifted in the concentration axial direction, respectively; then, other color concentration histograms are stretched in the concentration axial direction with the minimum concentration value used as reference, superimposed areas of concentration histograms by combinations of two color components after stretch-processing are calculated and derived, respectively, and based on the shift rate and stretch ratio which maximize the total of superimposed areas obtained respectively, each color component of each pixel of the color image data is converted.

Abstract: A photograph image-processing method, comprising steps of entering a film image for reading a film image by an image pickup device and generating color image data, generating concentration histograms for each of RGB color components from the color image data, relative-stretch-processing the concentration histograms, calculating and deriving a relative stretch ratio which maximizes the RGB superimposed area which has undergone relative-stretch processing, calculating and deriving an evaluation value of reliability with respect to the relative stretch ratio, calculating and deriving a correction stretch ratio based on the evaluation value, and conversion-processing RGB components of each pixel of the color image data in accordance with the correction stretch ratio.