JPS583059A - Picture inputting method of digitizer - Google Patents

Abstract

PURPOSE:To omit an unnecessary movement, to reduce capacity of a memory for storing a picture data, and also to execute processing at a high speed, by moving a heat to only a place in which a picture exists. CONSTITUTION:In an image sensor 1, 8 directions are selected from its center, and in each direction, 8 sections S1-S8 are decided. Subsequently, the algebraic sum of luminance of an element contained in each section is calculated. When the algebraic sum of luminance is taken in each section and a head 3 is moved in the direction where this algebraic sum is small, the head converned always tracks on a line. As a result, an unnecessary part having no picture information is not swept by the head 3, therefore, the processing time is quickened. Also, since a picture data is stored only in the direction where the head 3 moves, a memory is utilized effectively.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a digitizer input method.
Regarding the digitizer OWgI single-person method that requires less memory capacity to store 1KIi1m data and is capable of high-speed processing.To input figures into combinations and data, Digi-Eve is usually used to input feature points (orthogonal intersections, endpoints). 9 character positions, etc.) is used. Although this method is currently available and reliable, it requires a great deal of labor, and it often takes several days for large-format mechanical drawings. , there is also a method of inputting this into a combination. In this method, the II shape is expressed as a collection of points, so for example, 6 points/m (0,12
5 cabinets has a standard resolution of 7 axils), for example, A
When inputting version 0 (841EIIIIIIX 594mm), the required capacity is s2x 1o7 width and tp
Eat without being suggestive.

The present invention has been developed in consideration of these points, and has improved the image input method of the digitizer by moving the head only at the location where the image exists, thereby eliminating unnecessary movement and making effective use of memory. This has been achieved. below,
The present invention will be described in detail with reference to the drawings.

In general, the figures drawn on mechanical drawings require very fine rounding and high resolution, but on the other hand, the figures are not completely lost, so they can be read as many times as necessary. . The present invention was developed by taking advantage of the characteristics of drawings and considering drawings as a kind of model. FIG. 1 is a block diagram showing one embodiment of a digitizer for implementing the present invention.

, FlglK, 1 is an image sensor that receives iii* information and converts it into an electrical signal. As the cough image sensor, for example, a COD or a photodiode is used. Reference numeral 2 denotes a lens system that forms an image on the image plane 7y1.In the example shown in Fig. 1, the lens system 2 forms a real image of the partial area 5 on the image sensor 1 in the drawing , the straight line is captured in image 7Vf<,
The image sensor 1 and lens system 2 constitute a head that can freely move horizontally on the drawing 4. This head is numbered 3.

Figure 2 shows the direct tsto image captured by image sensor 1.
*It is a figure showing. Image 7/S1. The number of elements is 7×
Congratulations on the 7th. In the figure, the brightness of elements that completely overlap in a straight line is 1. The brightness of elements that overlap only partially is 1. The brightness of elements that do not overlap at all is defined as 2, and if we value each element according to this definition, we get something like the one shown in Figure S. Incidentally, the brightness distribution shown in Figure S is a calculated distribution, not an actual brightness distribution.

The actual luminance dynamic range is about 60 dB.

Next, as shown in FIG. 4, eight directions are selected from the center of the image shown in FIG. 3, and 80 to 3808 sections are determined for each direction. Then, calculate the algebraic sum of the luminances of the elements that meet in each section1
For example, sector 4.8. For, the elements are (2, 2,
i) (see Figure 5), so the algebraic sum is 2+2Φ1
■5, which is 4 combs.

85a and 85a indicate that the sum of the luminances is 0 and that the lines continue in this direction. Therefore, if the algebraic sum of the brightness is calculated for each sector and the head 5 is moved in the direction of -g of this algebraic sum, the head will always trace on the line. According to the method described above, the processing time can be shortened because the head device does not sweep unnecessary portions without image information. That is, high-speed operation is possible. Furthermore, since it is only necessary to store image data for the direction in which the head 3 moves, it is possible to effectively utilize the storage space. Therefore, since the printing capacity is small, it is possible to realize an inexpensive device as a whole.

FIG. 5 is a diagram illustrating a right right angle; - na turn using the method according to the present invention. Figure (a) In the FC, when the right right angle Poona is captured in the image area, there are seven points S in addition to the direction S8, where the right angle Puna is captured in the image area.
When 1 becomes 4, the brightness of the surrounding area becomes 9. Sara K
(b) and (a), the value of the moving direction sex scene S8 increases to 2, 4, and (
becomes 6, and sexy, n8. KO appears. Therefore, it can be argued that the O tracking after the point is a sector and the y82 direction, and at the same time the center is a corner.

゛Figure 6 shows the case where there are branches perpendicular to the direction of travel, and (4
), it is recognized that a straight line continues in the direction of S2 and S, and the center is the branch point.
I'll be damned.

By integrating this information, it is possible to create a W-weave with straight lines branching in a maximum of eight directions.
You can create input data that is similar to manually digitized data. Theory 1j mentioned above! For KTh, we took the 08 direction as an example from Sl to s8t as the traveling direction,
It is not necessarily limited to this, but it may be in four directions, including only the horizontal working direction, or it may be in eight or more directions.
Further, the number of elements constituting the image sensor is not limited to 47×7, and may be any other number.

Above are the details Km! As explained above, according to the 13iK of the present invention, it is possible to realize a digitizer image input method that moves the head only where the image exists, thereby eliminating unnecessary movement and effectively utilizing the memory. .

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the digitizer, FIGS. 2 to 4 are diagrams showing the configuration of an image sensor, and FIGS. 5 and 6 are diagrams showing the movement of the head. 1.-image sensor, 2.-lens system, 3.-head,
4 - drawing, 5 partial area, - straight line. Atsushi 5 (a) (b) (c) (d)'i! EJ6 figure (.) (b)

Claims (1)

[Claims] In a digitizer in which image data is detected by an image sensor by moving the head, (a) the image sensor is divided into several elements;
(b) A lens system is used on these divided image sensors to form a partial image of the figure to be gauged, and (e) the image sensor is divided into a plurality of sexy images extending radially from the center, and (b) Calculate the algebraic sum of the brightness of each sexy element of the elements included in each sexy sign, and (e) move the head along the direction where the algebraic sum is smallest to trace the figure weight.
It is characterized by detecting and storing graphical 0* signature points such as corners and branch schools due to changes in the direction of travel and the presence of sectors with low brightness outside the direction of travel, :/. How to create images without digitizing