Abstract: In an image encoding apparatus, an image area separating section determines for each pixel whether an input original image is a line image or a plane image and outputs separation information representing a type of image for each pixel. A line image position information encoding section encodes the separation information and outputs coded line image position information. A line image generating section extracts a line image from the input original image on the basis of the separation information. A line image color information quantizing section quantizes the line image and outputs quantized line image color information. A line image color information encoding section encodes the quantized line image color information and outputs coded line image color information. A plane image generating section extracts a plane image from the input original image on the basis of the separation information. A plane image quantizing section quantizes the plane image and outputs quantized plane image information.

Abstract: A line segment detector includes a first weighted average section which calculates the weighted average of pixel values in a first detecting window. The detector also includes a second weighted average section which calculates the weighted average of pixel values in a second detection window. A differentiating section calculates the absolute value of a difference between the outputs of the first and second weighted average sections. A dot detecting section calculates the average of absolute differential values between pixels in a vertical line direction, a horizontal line direction, and diagonal directions set in a detection window in an area containing the target pixel.

Abstract: An image memory is used for rotation of image data, and to shorten processing time. An image data rotating method uses an image memory capable of storing n×n pixels, and, for consecutive two pages in image data with the same size of a plurality of pages to be rotated, reads image data in a former page written in said image memory from each storage area at each address of said image memory in a predetermined order of reading according to the angle of 90 degrees to be rotated every k pixels for one line and every k lines for all lines; and writes image data in the latter page in the same storage area at the same address of said image memory in the predetermined order.

Abstract: In an image encoding apparatus, an image area separating section determines for each pixel whether an input original image is a line image or a plane image and outputs separation information representing a type of image for each pixel. A line image position information encoding section encodes the separation information and outputs coded line image position information. A line image generating section extracts a line image from the input original image on the basis of the separation information. A line image color information quantizing section quantizes the line image and outputs quantized line image color information. A line image color information encoding section encodes the quantized line image color information and outputs coded line image color information. A plane image generating section extracts a plane image from the input original image on the basis of the separation information. A plane image quantizing section quantizes the plane image and outputs quantized plane image information.

Abstract: A line segment detector includes first and second weighted average sections, differentiating section, dot detecting section, and comparing section. The first weighted average section calculates the weighted average of the pixel values in the first detection window set in the first neighboring area of a target pixel. The second weighted average section calculates the weighted average of the pixel values in the second detection window set in the second neighboring area point symmetrical with the first neighboring area. The differentiating section calculates the absolute value of a difference between outputs from the first and second weighted average sections. The dot detecting section calculates the average of the absolute values of differential values between pixels in a direction of a vertical line, horizontal line, left-to-right downward slanting line, or left-to-right upward slanting line within the third detection window set in an area containing the target pixel.