JP2968028B2 - Figure input processing method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a graphic input processing method for efficiently inputting graphic information constituting a map or various design drawings to a computer.

[Conventional technology]

As a typical method of the conventional graphic data acquisition technique, a drawing is pasted on a large tablet (digitizer), and each line segment constituting a graphic to be extracted is manually input. For this reason, the work efficiency was extremely poor and required a lot of time. In addition, when the size of the target drawing is a large size such as an A0 size, for example, a large tablet is used, and the operator has to stand and work hard.

[Problems to be solved by the invention]

On the other hand, a technique for automatically recognizing a figure in a drawing by a computer has also been studied. However, with the current recognition technology, the target figure is very limited, and all kinds of figure types necessary for practical use are available. Acquisition of figure data is almost impossible.

It is an object of the present invention to not only efficiently input graphic information (at high speed) to a computer, but also to significantly reduce the burden on an operator as compared with the prior art.

[Means for solving the problem]

FIG. 1 shows the principle processing mode of the present invention. Symbol 1 in the figure
Is a graphic type input process (first process) for inputting a graphic type. Reference numeral 2 denotes a tracking start partial figure designating process (second
), And designates a partial graphic constituting the input graphic. Reference numeral 3 denotes a process of extracting a partial figure group in the tracking direction (third process) and a process of calculating the degree of recognition certainty of each partial graphic (fourth process).
Step), a plurality of partial figures existing around the partial figure specified in the second step are extracted, and the extracted plurality of partial figures are input in the first step. The degree of certainty that can be a figure type is calculated. Reference numeral 4 denotes a process of highlighting a partial graphic, in which a partial graphic having a higher degree of certainty is sequentially highlighted. Reference numeral 5 indicates a display process (fifth process) for operator determination.

(Operation)

The present invention provides a graphic input processing method in which a computer is made to do what a computer can do, and an operator assists only when the computer is difficult to recognize, and information (graphic type) on a separately input graphic type is provided. Based on this, the degree of certainty, which is the figure of the figure type, is obtained, and the partial figure having the highest degree of certainty is displayed, thereby reducing the burden on the operator.

〔Example〕

The figures dealt with in the present invention are considered to be a drawing formed by a line drawing, and vector data obtained by simple vector approximation of a center line (center line) of the drawing, and coordinates of a pixel column forming the center line. Any expression format, such as value string data or chain-encoded data obtained by chain-encoding the pixel string, is targeted. Further, image information of the drawing input from the scanner device is also set as the target range.

Hereinafter, for the sake of simplicity of explanation and understanding, description will be made assuming that the basic graphic data to be handled is the above-described vectorized data (hereinafter referred to as vector data).

 Hereinafter, the processing mode will be described with reference to FIG.

Processing: If there are many graphic types to be recognized (extracted), they are displayed as a menu on a display, and the operator notifies the computer by picking one of the graphic types. Hereinafter, the following series of processing is executed on the assumption that the figure to be recognized is the figure type.

Processing: Next, a figure (drawing) to be recognized is displayed on a display, and the operator picks a partial figure constituting the figure to be extracted from the display using a mouse cursor or the like. The process of determining which partial graphic data the selected partial graphic corresponds to in the entire graphic data may use a function (known technology) normally provided in a normal graphic processing system. .

The computer performs the following processing with the picked partial graphic as the tracking start partial graphic.

Processing: First, a partial figure existing around the tracking start partial figure is extracted. As an example of this, a contour figure included in a normal map will be described as an example with reference to FIG. As shown in FIG. 2, it is assumed that the user wants to track in one direction of the tracking start partial figure. That is, a partial figure (group) existing around one end point a of the tracking start partial figure in FIG. 2 is extracted. In the example of FIG. 2, three partial figures A, B, and C can be extracted. The possibility that the three partial figures are connected as contour lines to the tracking start partial figure is calculated as the recognition certainty degree. In this contour line example, the directionality (the contour line does not bend at a steep angle or the like) with respect to the tracking start part figure or the property that the partial figure constituting the contour line is a curved curve with a large curve ( This is called monotonic curve characteristics), and contour lines in the map are extracted from various features, such as being formed by long lines to some extent, and comprehensive judgment is made to calculate the degree of certainty of each partial figure as a contour line. I do. Other graphic types also have various characteristics specific to the graphic type, and the certainty degree of the graphic type can be calculated using the various characteristics.
Let pA, pB, and pC be the recognition certainty degrees of the three partial figures obtained in this way. Also, 0.0 ≦ pA, pB, pC ≦ 1.0,
If the value is large, it is determined that there is a high possibility that the figure is a partial figure of the figure to be recognized.

Processing: Next, highlight colors are sequentially displayed on the display in order from the partial figure having the highest recognition certainty.

Here, if only one partial figure is extracted and its recognition certainty is high, or even if a plurality of partial figures are extracted, the recognition certainty of one partial figure is the same as that of another partial figure. If there is a sufficiently high possibility that the target figure can be a constituent element of the target figure type, the tracking is continued without highlighting the extracted partial figure on the display. That is, the process returns to the above processing.

Processing: The operator determines partial graphics to be sequentially highlighted in the processing, and (1) if it is a recognition target graphic, instructs the computer to that effect; It is determined that the object is a graphic, and the process returns to the above.

(2) If the highlighted color display graphic is not the target graphic, the process returns to display the partial graphic having the next highest recognition certainty. The processing is to be circulated.

(3) If there is no more partial figure to be traced in that direction, the process returns to the step of tracking the partial figure in the opposite direction.

(4) In the case of (3), when the tracking in the reverse direction has already been performed, it is assumed that the extraction of all the partial graphics constituting the recognition target graphic has been completed (completed), and the recognition processing ends.

In the example of FIG. 2, assuming that the partial figure A is selected by the operator in the processing, five partial figures (D, E, F, G, H) in the vicinity of the point b in FIG. Are extracted, the degree of recognition certainty is calculated as a contour line of each of the partial figures, and the above processings 1 to 3 are performed.

In each of the recognition degrees of the partial figures A, B, and C, for example, the recognition degree of the partial figure A is sufficiently high (pA≫pB, p
C) If the recognition accuracy of the partial figures B and C is extremely low (pB ~ 0.0, pC ~ 0.0), as described above, the processing is not performed, and the processing immediately returns to the processing in the vicinity of the point b. The extraction processing of the partial graphic and the calculation processing of the recognition certainty degree of the extracted partial graphic are performed. That is, in this case, the tracking / extraction of the next partial figure is continued without requiring the intervention of the tunnel.

The above procedure is repeated until the process (4) is completed (completed).

〔The invention's effect〕

As described above, according to the present invention, a computer sequentially designates partial figures constituting a recognized figure in descending order of the possibility of forming the recognized figure (recognition certainty degree). If continuous tracking is possible, the desired figure can be obtained without the need for operator intervention. That is,
Since the computer can do what the computer does and the operator only intervenes when the computer cannot be recognized, the input operation of the graphic data can be done efficiently (at high speed) and the conventional automatic recognition technology It is possible to handle all types of graphics without limiting the target graphics as shown in Fig. 2. Furthermore, the burden on the operator is greatly reduced compared to the conventional system, and the practical value is extremely high. .

Note that the recognition certainty at the time of tracking described above can be obtained for any type of figure. However, even if the accuracy, analysis, calculation, and calculation accuracy of the recognition certainty are low, all partial figures of the tracking candidate can be obtained. Extraction is possible, and these partial figures can be sequentially highlighted.

[Brief description of the drawings]

FIG. 1 is a view for explaining a processing procedure which is a principle processing mode of the present invention, and FIG. 2 is a view for explaining an example of recognizing a contour figure as an example of the figure recognition of the present invention. In the figure, 1 is a graphic type input process, 2 is a tracking start partial graphic designating process, 3 is a tracking direction partial graphic group extraction process and a recognition certainty calculation process of each partial graphic, and 4 is a partial graphic highlight color display process. Numerals 5 and 5 indicate a display process for operator judgment.

────────────────────────────────────────────────── ─── Continuation of the front page (56) References JP-A-63-6683 (JP, A) JP-A-1-14677 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06F 17/50

Claims (4)

(57) [Claims] 1. A graphic input processing method for displaying an input graphic on a display, comprising: a first step of inputting a graphic type to be recognized; and one of partial graphics constituting the input graphic. A second process in which the operator is instructed by the operator, a third process of extracting a partial graphic existing around the partial graphic specified in the second process, and a third process extracted in the third process. A fourth step of calculating the degree of recognition certainty representing the possibility that the partial figure can be the figure type input in the first step; and a partial figure having a high degree of recognition certainty calculated in the fourth step. A fifth process of sequentially highlighting on the display, and a sixth process in which, when the partial graphic sequentially displayed in the fifth process is the graphic type, the operator is instructed to that effect, Having said input A third step to a sixth step until all components of the input graphic are extracted; and inputting information on the input graphic to a computer. 2. A method according to claim 3, wherein when the maximum value of the degree of certainty of recognition of the partial graphic calculated in the fourth step is equal to or greater than a predetermined threshold value, the partial graphic corresponding to said maximum value is set as the designated partial graphic in the third step. 2. The graphic input processing method according to claim 1, wherein the fifth step is executed when the certainty degree is less than the threshold value. 3. The method according to claim 1, wherein the third step to the sixth step are performed in the sixth step.
3. The graphic input processing method according to claim 1, further comprising a step of requesting and accepting a judgment as to whether or not to end the repetition of the step. 4. When the judgment as a component of the figure to be recognized is denied in the sixth step, the fifth step is repeated,
4. The graphic input processing according to claim 1, wherein the fifth step displays one partial graphic at a time and displays a partial graphic having the second highest recognition certainty each time it is repeated. Method.