Image analysis – Histogram processing – For setting a threshold
Reexamination Certificate
1999-08-11
2003-04-15
Do, Anh Hong (Department: 2624)
Image analysis
Histogram processing
For setting a threshold
C382S270000
Reexamination Certificate
active
06549655
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a binarization method of binarizing an image by extracting lightness (brightness, density) as a feature amount from the image, and an image processing system using the method.
2. Description of the Related Art
In extraction processing intended for extracting an object from an image, it is important to discriminate areas contained in the image from each other. In a method of discriminating the areas, it is common practice to discriminate the image as a partial image having uniform feature amounts of lightness, color, texture, and the like.
The method of recognizing the areas with the lightness as the feature amount is already used in various industrial fields; for example, it is applied to determination as to whether or not the title lettering of a product exists, determination of the front or back of a part, inspection on parts for loss, inspection for dirt and flaw, and the like.
In a system for such purposes, an image of the object such as a product is picked up using an image pickup machine such as a camera and the picked-up image is input to an image processing system as an analog signal comprising Y values representing the lightness for each pixel, then the analog Y values input are converted into digital Y values, which then are stored in a storage unit. The image processing system displays the original image on a display unit such as a CRT based on the stored information.
By the way, to execute the determination and inspection as described above, normally a binary image is used, thus hitherto it has been common practice to directly enter a numeric value for the displayed image through an input unit from the outside, thereby binarizing the displayed image with the entered numeric value as a threshold value.
However, in the method as described above, the lightness (brightness) to be extracted is entered in a numeric value, thus it is sensuously difficult to associate the entered numeric value with the lightness corresponding thereto. This results in extreme difficulty particularly if the lightness difference in the original image is small.
SUMMARY OF THE INVENTION
It is therefore an object of the invention to provide an image binarization method wherein a pixel to be extracted on a screen can be directly selected sensuously easily without the need for setting a threshold value in a numeric value, and to provide an image processing system using the method.
According to a first aspect of the invention, there is provided an image binarization method comprising the steps of setting a threshold value range for a value concerning a Y value representing brightness of a pixel selected from an image and extracting a pixel with the value concerning the Y value in the image within the threshold value range, wherein when a pixel is selected in the image, the threshold value range is set in a range of adding a predetermined value to the value concerning the Y value of the selected pixel and/or subtracting the predetermined value therefrom, and wherein when after the pixel is selected, a pixel different from that pixel is selected, if the value concerning the Y value of the selected pixel is greater than the value concerning the Y value of the pixel for which the threshold value range is set, between the predetermined value added to the value concerning the Y value of the selected pixel and a lower limit value in the threshold value range, a new threshold value range is set, or if the value concerning the Y value of the selected pixel is smaller than the value concerning the Y value of the pixel for which the threshold value range is set, between the predetermined value subtracted from the value concerning the Y value of the selected pixel and an upper limit value in the threshold value range, a new threshold value range is set.
According to a second aspect of the invention, there is provided an image binarization method comprising the steps of setting a threshold value range for a value concerning a Y value representing brightness of a pixel selected from an image and extracting a pixel with the value concerning the Y value in the image within the threshold value range, wherein when an unextracted pixel is selected, the threshold value range is set in a range of adding a predetermined value to the value concerning the Y value of the selected unextracted pixel and/or subtracting the predetermined value therefrom, and wherein when an already extracted pixel is selected, if the value concerning the Y value of each of one or more unextracted pixels of pixels in a predetermined range containing the already extracted pixel selected is greater than an upper limit value in the threshold value range, using the value concerning the Y value of the unextracted pixel with the minimum difference between the value concerning the Y value of each of one or more unextracted pixels and the upper limit value in the threshold value range as an upper limit value, a new threshold value range is set, or if the value concerning the Y value of each of one or more unextracted pixels of pixels in a predetermined range containing the already extracted pixel selected is smaller than a lower limit value in the threshold value range, using the value concerning the Y value of the unextracted pixel with the minimum difference between the value concerning the Y value of each of one or more unextracted pixels and the lower limit value in the threshold value range as a lower limit value, a new threshold value range is set.
According to a third aspect of the invention, there is provided an image processing system for setting a threshold value range for a value concerning a Y value representing brightness of a pixel selected from an image and extracting and displaying a pixel with the value concerning the Y value in the image within the threshold value range, the image processing system comprising: first threshold value setting means, when a pixel is selected in the image, for setting the threshold value range in a range of adding a predetermined value to the value concerning the Y value of the selected pixel and/or subtracting the predetermined value therefrom; and second threshold value setting means, when after the pixel is selected, a pixel different from that pixel is selected, if the value concerning the Y value of the selected pixel is greater than the value concerning the Y value of the pixel for which the threshold value range is set, between the predetermined value added to the value concerning the Y value of the selected pixel and a lower limit value in the threshold value range, the second threshold value setting means for setting a new threshold value range, or if the value concerning the Y value of the selected pixel is smaller than the value concerning the Y value of the pixel for which the threshold value range is set, between the predetermined value subtracted from the value concerning the Y value of the selected pixel and an upper limit value in the threshold value range, second threshold value setting means for setting a new threshold value range.
According to a fourth aspect of the invention, there is provided an image processing system for setting a threshold value range for a value concerning a Y value representing brightness of a pixel selected from an image and extracting and displaying a pixel with the value concerning the Y value in the image within the threshold value range, the image processing system comprising: first threshold value setting means, when an unextracted pixel is selected, for setting the threshold value range in a range of adding a predetermined value to the value concerning the Y value of the selected unextracted pixel and/or subtracting the predetermined value therefrom; and second threshold value setting means, when an already extracted pixel is selected, if the value concerning the Y value of each of one or more unextracted pixels of pixels in a predetermined range containing the already extracted pixel selected is greater than an upper limit value in the threshold value range, the second thresho
Ikushima Yasuhisa
Saeki Kazuhito
LandOfFree
Image binarization method and image processing system using... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image binarization method and image processing system using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image binarization method and image processing system using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3021523