JPH06105973B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of application) The present invention is an image processing apparatus for processing an image obtained from a TV camera and measuring its characteristic amount, which can be used for a general industrial work handling / processing apparatus. Regarding

(Prior Art) As such an image processing device, for example, a multi-window type image information masking device disclosed in JP-A-58-156273 is disclosed, and such a multi-window type image information device is known as a commercial product. There is. These are generally known as devices that limit the processing area of the image obtained from the camera by a window, that is, a small window for measurement, but in these, the position of the window is fixed and the processing is performed. There were considerable restrictions when the position of the desired image was indefinite.

(Problems to be Solved by the Invention) An object of the present invention is to solve this problem and realize higher-speed image processing. That is, in order to perform more accurate image processing, it is necessary to obtain an image composed of more pixels. However, the large number of pixels necessitates more processing, which naturally increases the processing time. It is an object of the present invention to provide a faster and more accurate image processing apparatus by performing more accurate image processing only on an image of interest and by performing high-speed image processing.

(Means for Solving the Problems) Therefore, in the present invention, in the image processing device, when the image processing is performed, the digital original image is reduced, and the obtained reduced image is subjected to the image processing to obtain the image of interest. The image processing apparatus is characterized in that after the area is detected, the image processing is performed again only on the image area of interest on the digital original image before reduction.

(Example) FIG. 1 shows a main configuration of an image processing apparatus of the present invention, and FIG. 2 shows a flow chart of an outline of processing of this apparatus. The image obtained from the TV camera (a) is converted from an analog amount into a digital amount by the A / D conversion means (b) and stored in the original image memory means (c). (For example, divide the image into 512 x 512 points and use 1 BYT of the original image memory means for each point.
E has a value according to the brightness. ) Next, the digital original image in the original image memory means (c) is converted into a reduced image (256 × 256 or 128 × 128 points) in the image reduction processing section (d), and this reduced image is reduced image memory means. It is stored in (e).

The image reduction processing unit reduces the original image (A) to a reduced image 1 (B) or a reduced image 2 (C) as shown in FIG. 3, which is shown in FIG. like,
Of all the pixels of the original image, the shaded pixels in FIGS. 4 (B) and (C) (that is, every 1 pixel vertically and horizontally when the reduction ratio is 1/4, and vertically and horizontally when the reduction ratio is 1/16) (Every 3 pixels)
Only the effective pixels are continuously rearranged on the reduced image memory. By this, the digital original image may be effectively reduced on the reduced image memory means.

Next, the reduced image attention area detection unit (f) performs various known image processing on the reduced image to detect the image to be noticed, and obtains the area occupied by the attention image as the address on the reduced image memory means. Based on the detected area of interest on the reduced image memory means, the area of interest on the original image is set, and the area of interest on the digital original image stored in the original image memory means (c) is set. Only the region of interest on the digital original image is processed. That is, in the original image attention area detection unit (f ′), the address on the reduced image memory means (e) indicating the reduced image attention area is multiplied by the reduction ratio to obtain the original image memory means (c). It is converted into an address indicating the image area of interest (FIG. 5). Finally, in the region-of-interest image processing unit (g), in the digital original image on the original image memory means obtained by the original image region-of-interest detecting unit (f ′), only the pixels corresponding to the address indicating the region of interest image are extracted. Image processing is performed by a known method.

(Advantageous Effects of the Invention) According to the present invention, it takes a long processing time to process an image having a large number of pixels such as 512 × 512 pixels. It takes only 1 processing time.

Furthermore, noise and images that do not require processing are effectively erased, so that it is possible to considerably reduce the possibility of erroneous processing. Even if the image area of interest is indefinite, the image area of interest on the original image is detected in a short time on the reduced image, and then image processing is performed only on the image area of interest on this original image. Therefore, it is more reliable and versatile than the conventional method of processing only the fixed area.

As described above, the present invention can realize accurate image processing with few errors at high speed.

[Brief description of drawings]

FIG. 1 is a claim diagram showing a main configuration of an image processing apparatus of the present invention, FIG. 2 is a schematic flow chart of the apparatus shown in FIG. 1, and FIG. 3 is an original image (A) and a reduced image 1 (B). And FIG. 4 is a diagram showing a reduction relationship with the reduced image 2 (C), and FIG.
FIG. 5 is an explanatory diagram showing a state in which effective pixels, that is, shaded portions, are extracted from the original image pixels so as to be the reduced image 1 (B) and the reduced image 2 (C) of FIG. 5, and FIG. 5 is the original image attention area detection unit of FIG. It is explanatory drawing which shows the effect | action of. a: TV camera b: A / D conversion means c: Original image memory means d: Image reduction processing section e: Reduced image memory means f: Reduced image focused area detection section f ': Original image focused Region detection unit g ... Region of interest image processing unit

Claims (1)

[Claims] 1. A TV camera for outputting image information, an A / D conversion means for converting the image information of the TV camera into a digital amount, and an original for storing the output of the A / D conversion means as a digital original image. An image memory unit, an image reduction processing unit for converting the digital original image stored in the original image memory unit into a reduced image, a reduced image memory unit for storing the output of the image reduction processing unit as a reduced image, A reduced image attention area detection unit for processing the reduced image stored in the reduced image memory means to detect an area of an image to be noticed, that is, a reduced image attention area, and the reduced image attention detected by the reduced image attention area detection unit An original image attention area detection unit that sets an attention area in the digital original image stored in the original image memory means from an area, that is, an original image attention area, and the original image attention area detection unit An image processing apparatus, comprising: a region-of-interest image processing unit that processes an image processing region of the digital original image designated by the region of interest of the original image.