WO2016145581A1

WO2016145581A1 - Method for generating precise pattern according to hand-drawn graph

Info

Publication number: WO2016145581A1
Application number: PCT/CN2015/074250
Authority: WO
Inventors: 王浩屹; 王建新
Original assignee: 王浩屹; 王建新
Priority date: 2015-03-13
Filing date: 2015-03-13
Publication date: 2016-09-22

Abstract

A method for generating a precise pattern according to a hand-drawn graph, comprising the following steps: S1: recognizing at least one graphic element of a hand-drawn graph; S2: recognizing identification information corresponding to the graphic element; and S3: generating a precise pattern corresponding to the graphic element according to the identification information. In the method, by means of recognizing identification information corresponding to a graphic element, it is realized that the graphic element on a hand-drawn graph is converted to generate a corresponding precise pattern, so that the time for a byte input apparatus, such as a keyboard, needing to be used in the prior art to input the graphic element is avoided. The method has advantages of being high in working efficiency, saving time costs, etc.

Description

Method for generating accurate patterns based on hand drawings

Technical field

The present invention relates to a graphics processing method, and more particularly to a method of generating an accurate pattern based on a hand drawing.

Background technique

With the development of science and technology, it is very common for people to use computer to draw. For example, using AutoCAD, Zhongwang CAD and other vector drawing systems for architectural design drawing, mold design drawing, mechanical design drawing and so on. The use of these vector drawing systems has brought great convenience to people, but these drawing systems still have many shortcomings: these drawing systems usually require the user to input the specific graphic element requirements that the computer can handle, and use the keyboard on the drawing area. Byte input device.

In fact, hand-painting is the most primitive and basic way for humans to express and think about graphics. Modern use of computers to draw vector paper through the mouse and keyboard is just a helpless choice when humans use computers. The reason why hand-drawn drawing tools that are not suitable for modern computers are drawn is that human hands-free drawing is impossible to draw graphics accurately, but computers can only process vector graphics by processing precise data.

From the computer graphics point of view: for human hand-drawn, the computer has always been processed as a bitmap image, but the use of computer processing vector paper has long been a tool for most industries such as engineering design, mechanical design.

Summary of the invention

The technical problem to be solved by the present invention is to provide a method for directly generating an accurate pattern according to a hand drawing.

The technical solution adopted by the present invention to solve the technical problem thereof is: constructing a method for generating an accurate pattern according to a hand drawing, comprising the following steps:

S1: identifying at least one graphic element of the hand drawing;

S2: identifying identification information corresponding to the graphic element;

S3: Generate an accurate pattern corresponding to the graphic element according to the identification information.

In the method of generating an accurate pattern according to the hand drawing of the present invention, the identification information includes instruction flag information; in the step S3, the graphic element is processed according to the instruction flag information.

In the method for generating an accurate pattern according to the hand drawing of the present invention, the instruction mark information includes one or more of a delete instruction, a modification instruction, an addition instruction, and a query instruction.

In the method for generating an accurate pattern according to the hand drawing of the present invention, the identification information includes graphic type information and graphic annotation information;

The step S3 includes:

S3-1: determining whether the graphic type information is included in the identification information, if yes, executing step S3-2, otherwise performing step S3-3;

S3-2: determining a graphic type according to the graphic type information, and generating a standard graphic of the graphic of the type;

S3-3: Determine a graphic type corresponding to the graphic element according to the graphic element, and generate a standard graphic of the graphic of the type.

In the method for generating an accurate pattern according to the hand drawing of the present invention, the step S3-3 includes the following steps:

S3-3-1: a first endpoint, a second endpoint, and at least one sampling point between the first endpoint and the second endpoint in the identified graphical element;

S3-3-2: identifying, according to the first endpoint, the second endpoint, and the sampling point, hand-drawn graphic information of the graphic element between the first endpoint and the second endpoint, and according to the The hand drawn graphic information gets the graphic type.

In the method for generating an accurate pattern according to the hand drawing of the present invention, in the step S3-3-2, connecting the first end point and the second end point, determining the sampling point and the first end point a distance relationship between the second endpoint connection lines, the hand-drawn graphics information is obtained to determine the type of the graphic between the first endpoint and the second endpoint; or

Taking the first endpoint as a starting point, continuously calculating a function relationship of the next sampling point until the second endpoint, according to the obtained function relationship, obtaining the hand-drawn graphic information to determine the first end The type of graphics between the point and the second endpoint.

In the method for generating an accurate pattern according to the hand drawing of the present invention, the method further includes a canvas setting step:

A drawing area is set for the user to draw the hand drawing in the drawing area through an input device.

In the method for generating an accurate pattern according to the hand drawing of the present invention, by identifying the initial input point of the input device in the drawing area and the input trajectory process, determining that the first end point, the second end point, and Sampling point.

In the method for generating an accurate pattern according to the hand drawing of the present invention, the step S3 further includes:

S3-4: determining whether there is graphic annotation information in the identification information, if yes, executing step S3-5, otherwise performing step S3-6;

S3-5: The standard pattern obtained by the step S3-2 or S3-3 is staked according to the graphic annotation information, and the accurate pattern is generated;

S3-6: Perform fuzzy calculation according to the position and shape of the brush cover canvas in the drawing area according to the graphic element, and stake out the standard graphic obtained in step S3-2 or S3-3 to generate the accurate pattern.

In the method for generating an accurate pattern according to the hand drawing of the present invention, the step S3-5 includes the following steps:

S3-5-1: determining whether the graphic annotation information includes unit information, if yes, executing step S3-5-2, otherwise performing step S3-5-3;

S3-5-2: generating the accurate pattern according to the annotation data and the unit information in the graphic annotation information;

S3-5-3: determining unit information according to the annotation data in the graphic annotation information, combining the scale of the hand drawing, and/or the user setting ratio, and/or a common unit, and according to the annotation data and The unit information is staked to generate the exact pattern.

In the method for generating an accurate pattern according to the hand drawing of the present invention, in the steps S1 and S2, graphic elements and/or data elements are obtained according to the identified graphic elements;

In the step S3, it is determined whether the graphic element and/or the data element obtains a valid component group; if a valid component group is obtained, a basic element group is generated, and the basic element group data is generated according to the obtained basic element group data. Precise pattern.

In the method for generating an accurate pattern according to the hand drawing of the present invention, in the steps S1, S2, identifying an element related to the identification information;

Determining a reference element based on the identified element;

Identifying graphic type information and graphic annotation information in the identification information;

The identified element, the reference element, the graphic type information, and the graphic annotation information constitute the graphic element and/or the data element.

The present invention has the following beneficial effects: the present invention realizes converting the graphic elements on the hand drawing to generate corresponding accurate patterns by recognizing the corresponding identification information of the graphic elements, thereby avoiding the need to additionally use the keyboard and the like in the prior art. The input tool re-enters the time to draw the accurate pattern, which has the advantages of high work efficiency, time saving cost, and more in line with the way people express and think about graphics.

DRAWINGS

The present invention will be further described below in conjunction with the accompanying drawings and embodiments, in which:

1 is a flow chart showing an embodiment of a method for generating an accurate pattern according to a hand drawing according to the present invention;

2 is a flow chart showing the steps of generating an accurate pattern in the method for generating an accurate pattern according to the hand drawing of the present invention.

detailed description

As shown in FIG. 1, it is a flowchart of an embodiment of a method for generating an accurate pattern according to the present invention, by identifying at least one graphic element of a hand drawing (step S1); and then identifying corresponding identification information in the graphic element. (Step S2); Then, an accurate pattern corresponding to the graphic element is generated based on the identification information (step S3). Through this method, the contents of the selected range of the original drawing can be added, deleted, changed, checked, etc., so as to facilitate the purpose of modification; or a new accurate pattern can be generated according to the newly drawn hand drawing, and the standard is generated. The model is thus represented as a standard drawing.

Wherein, the graphical element is an element of the hand-drawn graphic (one or more combinations of points, lines, faces) used for avatar, and may include graphics and/or one or more identification information about the graphic. For example, the graphic element contains three parts: (1) an inaccurate line (graphic) of hand-drawn, (2) a length dimension of the line (graphic) of 3m (identification information), and (3) the line. An endpoint coordinate (identification information), in the above three parts, (1) is a graphic, (2) and (3) are identification information, which can be combined into one graphic element; if only the above (1) can also be used as a graphic element Although it is not precise, it can be identified by other methods; if only (2) and (3), it can also be used as a graphic element, only need to indicate that it represents a straight line, and can also be correctly identified; if only (2) Although it can also be used as a graphic element, the expression is a 3m long line, but the exact position of the line needs to be determined by other methods; if only (3) as a graphic element, only one coordinate identification information, it can only be accurate. Position the position of a point instead of a precise line.

When generating the accurate pattern, it is first determined whether the identification information includes the instruction mark information; if so, the graphic element is processed according to the instruction mark information. The instruction tag information includes one or more of a delete instruction, a modification instruction, an addition instruction, and a query instruction.

Meanwhile, as shown in FIG. 2, it can also be determined whether there is graphics type information in the identification information (step S3-1), if yes, step S3-2 is performed, otherwise step S3-3 is performed;

S3-3: Judging the corresponding graphic type according to the graphic of the graphic element, and generating a standard graphic of the graphic of the type. a first endpoint in the identified graphical element, a second endpoint, and at least one sample point between the first endpoint and the second endpoint; identifying the first end based on the first endpoint, the second endpoint, and the sampling point The hand-drawn graphic information of the graphic element between the point and the second end point, and the graphic type is obtained according to the hand-drawn graphic information. Among them, the standard graphics that generate the type of graphics can be generated by various geometric methods, for example, by straight line drawing, angle drawing, various geometric drawing, and the like.

The identification of the first endpoint, the second endpoint, and the sampling point can be determined by identifying the initial input point of the input device in the drawing area and the input trajectory process.

Then, S3-4: determining whether there is graphic annotation information in the identification information, if yes, executing step S3-5, otherwise performing step S3-6;

S3-5: according to the graphic annotation information, stake out the graphics obtained in step S3-2 or S3-3 to generate an accurate pattern;

S3-6: Perform fuzzy calculation according to the position and shape of the brush covering the canvas in the drawing area, and stake out the image obtained in step S3-2 or S3-3 to generate an accurate pattern.

In this embodiment, the method includes a step of setting a canvas, a step of drawing a graphic element, a step of identifying a graphic element, and a step of generating an accurate pattern. It can be understood that the above steps can be deleted or sequentially adjusted as needed.

In the step of setting the canvas, it is equivalent to setting an area for drawing, for the user to draw the hand drawing. The user can perform hand-drawing through various input devices of the hand drawing to form a hand drawing that needs to identify and generate an accurate pattern.

In this embodiment, the hand drawing may be using a mouse, a keyboard, an electronic tablet, or a mobile phone with a touch screen, a Pad device, a scanner or a camera image capturing device, and the like as an input device, and the drawing may be drawn. A combination of graphs or graphics such as dots, lines, and faces, inputting images and information processed by a computer. When the hand drawing is drawn using paper-based media and then input to the computer using a scanner or camera-type image capture device, it can be considered as a bitmap image; when using an electronic tablet, or a mobile phone with a touch screen, or a Pad device, The information entered may be more than just a bitmap image, and may also have hand-drawn sequences, starting point, end point, coordinates of key points, and the like. When using the mouse or keyboard input, the hand drawing also has additional information entered. The hand drawings are all inaccurate. At the current level, it is difficult for people to draw a figure with precise dimensions.

First, set the canvas size, which is the range that can actually be used for hand-drawn drawing. Such as the size of a piece of paper, the size of the screen, and so on. Determine the canvas range, select the need to add, delete, modify, query the model or model part, the model or model part is presented on the 2D display device in the form of drawings or output on the paper type media.

Then, set the origin of the canvas. The user can set the specified point as the origin through the input device. When the user does not specify, the horizontal coordinate of the origin is set to select the horizontal coordinate of the leftmost point of the canvas range, and the vertical coordinate is set to the vertical coordinate of the lowest point of the selection range.

Then, set the canvas direction. The user can set the specified direction as the direction of the canvas through the input device, so as to facilitate the subsequent reference of the driving graphic element. It can be understood that when the original image is modified by the method, the specified direction of the original image can also be directly called.

Then, set the canvas ratio. The user can set the scale by inputting the device; you can also set the scale using the scaled reference graphic element; in the case of the original image, you can also use the scale used to draw the original image in the canvas.

The reference graphic element of the scale may be: when setting the canvas scale, there are multiple graphic elements in the figure, and only one of them is used as a basis for calculating the set proportion, and the graphic element is the reference graphic element of the scale. The scaled reference graphic element can be a graphic element used to generate an accurate pattern in a hand drawing, or a graphic element specifically used to set a scale.

If you are drawing a new image without the original image, the new image does not have a scale. Here is a way to set the exact scale using the hand-drawn graphic elements: (1) If the scaled reference graphic element is set, The scaled reference graphic element is calculated. If the scaled reference graphic element is not set, the scale of the covered canvas is found in all the graphic elements of the hand drawing, and the complete graphic annotation information in the graphic element is used as the reference of the scale. Graphical elements; (2) the actual size represented by the graphic pointed to by the graphic annotation information in the scaled reference graphic element; the ratio of the meaning of the labeled text (labeled data or labeled data with units); (3) If the common scale value library is preset, the scale value obtained in the previous step is compared with the scale value in the common scale value library, generally taking a common scale value close to and smaller than the scale value obtained in the previous step, if the common ratio is not set. Value, go directly to the next step; (4) set the scale value obtained in the previous step to the scale of this canvas. The methods involved in steps (1) through (4) can be used as a set of technical solutions.

Then, the user can set different attribute parameters corresponding to different graphics and different identification information, and different attribute parameters may include colors, layers, internal attribute parameters, etc., thereby facilitating hand-drawn graphics.

At the same time, set the exact pattern corresponding to the graphic elements, as well as the exact pattern lofting to become the model, and the method of the model to generate the drawing. For example, the line with red (different color is parameter) is the center line of a building component wall. The thickness of this wall is 200mm and the height is 3000mm. The exact pattern of the red pair will be offset by 100mm on both sides. Then the seal (precision pattern) stretches the cube (model) of 3000mm to build the model, and then outputs the plane of the wall member in the form of a building plan; at the same time, set the point of black (different colors as parameters) as the circular cross in the building component. The center of the column with a cross section of 200 mm and the height of the column is 3000 mm. When the column is generated, the cylinder with the height of 3000 mm is retracted from the circle with the black point of the circle of 200 mm (accurate pattern). ).

In order to be able to lay out a canvas suitable for hand-drawing, it is possible to draw on the canvas some graphics that are easily discernible and used to position the hand-drawn images. These graphics can pre-arrange some positioning points, lines and faces on the canvas. When people draw on the canvas, they only need to explain the relative positional relationship of the hand-drawn graphics to these positioning points, lines and faces. The system will automatically follow the coordinate geometry principle. Calculate the vector graphics information that the hand-drawn graphic elements need to express. These graphics for identification and positioning include: original graphics and meshes or custom positioning graphics.

Draw a grid in the canvas: The first step is to determine the actual spacing of the grid. If the actual spacing is defined, it is determined by defining the actual spacing. If the actual spacing is not defined, the actual spacing of the grid is determined according to the scale of the graph. The actual spacing is the actual size represented by the actual spacing of the grid. Each scale will infer the common modulus values for this scale drawing graph. Inferred methods, for example, between 1:100 and 1:500, will be drawn a plan of a building, China's main modulus is 100mm, 300mm and so on. If the ratio is less than 1:500, the general plan plan is drawn. The modulus used in the plan is generally an integral multiple of 1 m or an integral multiple of 1.5 meters. It can be seen that the actual spacing of the grid represented by the scale may be: greater than 5:1, representing 1 mm. Less than 5:1 is greater than 1:100, representing 10mm or 20mm or 50mm. Less than 1:100 is greater than 1:500 , representing an integer multiple of 300mm or 100mm, less than 1:500, representing 1000mm or 1500mm or 3000mm. In the second step, the grid spacing is determined, and the grid spacing is generally between 10 mm and 3 mm. The grid presented on the canvas is generally sized to be recognizable by human vision.

If you need to draw the original drawing on the canvas: select the model or model part that needs to be processed, you need to preset the style of the model to generate the drawing, including line type, color, layer and so on.

For the convenience of drawing, a preferred solution is to use the corresponding method for generating the drawing according to the correspondence relationship, and correspondingly output the same correspondingly. For example, the wall is displayed in the plan as a thick solid rectangle in the plan view, and the red line represents the center line of the wall. Although there is no center line of the wall in the original image, the center is still needed when drawing the original image in the canvas. The line is calculated and drawn on the canvas in the same red color. This is just a setting for the convenience of drawing. If there is no need to display it, there is no need to output it.

Draw a grid to determine the grid origin. If the grid origin is specified, the specified point is used as the origin. If not specified, the previously determined canvas origin is used as the origin.

Draw a horizontal and vertical grid, from the grid origin, to the canvas horizontally and vertically to the canvas across the canvas.

After the above-described setting canvasing step is completed, the user can perform a step of drawing a graphic element on the canvas by using an input device, thereby forming a hand drawing. Of course, setting the canvas step is not necessary. The setting content of the canvas can be directly called for the standard format template. The timing of the canvas setting can also be drawn after the drawing is completed. .

The drawing of the graphic steps may include drawing a graphic of the graphic element, drawing identification information of the graphic element, and the like. Drawing graphic elements should be based on the following principles: (1) All graphic elements should have preset parameters for recall when they are recognized; (2) the position and shape description of the graphic elements should satisfy the geometric elements. A component group; (3) Each feature value in the valid component group must have a valid input method, otherwise the graphic element will not be recognized.

The input method described in the above principle (3) means: (1) directly inputting elements using a byte input device such as a keyboard; (2) inputting elements using graphic annotation information; (3) using a grid or original graphics or custom positioning Graphical input elements; (4) using the coordinates of the hand-drawn contacts fed back by the device, and then calculating the calculated features. There is a certain priority in the selection method of the above input method. Generally speaking, (1)>(2)>(3)>(4), if one element is input by multiple modes, it is input with a high priority. The elements are subject to change. It can be understood that the input method can also be combined by the above several input methods, and the priority is related to the input method of all the elements in the effective constituent element group. For example, at point A is a point on the grid, and now point B is determined from point A to 45 degrees 3m to the upper left, and the angle is entered using the angle annotation method of the graphic annotation information (2), and the contact feedback is used for 3m. The method input is (4), so the priority of point B as a separate graphic element is related to the priority of the above two input methods, and different working modes can calculate different priorities. The significance of determining the priority of an element is that when the various elements obtained by the graphic can be composed of a plurality of effective constituent elements, and the basic element groups obtained by the effective constituent elements are not necessarily the same or have a certain error, this When it is judged by priority, which valid component group is adopted.

As can be appreciated, an element is an element that describes a graphic. For example, a point on a plane is described by two coordinates: X and Y. The two coordinates of X and Y are the elements of this point; a segment of a line whose graphic elements are two endpoints (A and B), or two endpoints. Four coordinates (XA, YA and XB, YB). A graphic itself can be used as a graphic element of another graphic.

Graphical elements and data elements can be elements that describe a graphic. To be processed inside the computer, the elements must be digitized. The data elements are a form of data representation of the graphic elements, such as: a segment of a line, the graphic elements are two The endpoints (A and B), or their data elements are the four coordinates of the two endpoints (XA, YA, and XB, YB). It can be further understood that the elements of the line segment AB are composed of the graphic element A and the data element XB, YB composition. By identifying the graphical elements, graphical elements and/or data elements corresponding to the graphical elements are obtained.

The basic feature group can be a group of the most primitive data elements that the computer uses to process a graph. All the feature values in the group need to be processed, and the graph can be processed. For example, the computer needs to process a point. In general, the computer processes the X and Y coordinates of the point; the computer processes a line segment, and the computer processes the four coordinates of the two endpoints of the line segment (XA, YA, and XB, YB); There are also special cases, if the four coordinates are input to the computer, but one of the coordinates (XA, YA) and one vector (angle O and length L), then the computer will be input by the four The elements, through the calculation of the trigonometric function, calculate XB and YB, because (XA, YA, and XB, YB) are the data elements (basic feature groups) that the computer actually processes. Of course, depending on the computer program, the basic feature groups may be different, depending on the application, some programs may use vectors, some may use three-dimensional coordinates, and so on.

A component group can be a group of elements used in a computer to process a basic feature group that can be used to derive a graph without using other elements. The component group may not be a basic feature group, but only through the elements in the component group. Calculation, you can get the basic feature group. Of course, the component group can be directly a basic feature group, but when the basic feature group is lacking, it is entirely possible to use other features to derive the basic feature group. For example, a circle can be determined by (a) the center and radius, or by (b) any three points on the circle, or even by (c) the center of the circle and the area of the circle, above (a) (b) (c) are all constituent elements, and the elements in any one of them are complete, and this figure can be accurately constructed.

A valid component group is a group of component groups that can be processed into a basic feature group. All of the elements must be valid. In the present embodiment, whether or not the effective constituent element group is obtained by judging the obtained graphic element and/or data element; if the effective constituent element group is obtained, the basic element group is generated, and an accurate pattern is generated based on the obtained basic element group data.

The graphic types of the drawn graphic elements include, but are not limited to, points, lines, faces, and the like. Among them, the tap type classification includes a single point, an endpoint, an intersection point, and the like. A single point can identify the desired point with a figure of no more than 5 mm; both ends of the line are treated as two points as endpoints; points where several lines intersect are considered intersections.

Lines include lines, polylines, arcs, curves, and more. Each of the lines has its own set of multiple constituent elements, such as an arc with a clear arc length, a radius of the center of the circle, a third point, or a component group composed of other elements. The curve may be a plurality of arcs or a plurality of sampling points to form a component group, or may be a component group of a smooth curve composed of points that are not on a straight line or on the same arc. It can be understood that when the line does not have a clear schematic type information, it needs to be judged when it is recognized.

The face includes a face contour and a face fill, wherein the face contour can be surrounded by lines that meet end to end, and the face fill can be done by introducing a fill line or drawing a fill line.

It can be understood that the user can also customize more graphic elements according to work habits, and these graphic elements may not be limited to points, lines, faces and identification information. Then set up the component groups for the custom graphic elements and the method for generating an accurate pattern from each of the component groups.

An accurate pattern is a graphic that is used to stake out by identifying and calculating a process graphic. For example, when drawing a line, give this line a precise length of 3m, and determine the coordinates of the two end points of the line, set this line to be the center line of the wall. At this time, the 3m long line (that is, the exact pattern) is recognized. After the line is laid out as a model of a wall, this line is not necessarily presented in the form of a line. Instead, the line is centered and offset by the preset wall thickness. After that, it is presented as the final result in the form of a wall outline. In this embodiment, the accurate pattern may be a 3 m long line that is not staked out as a wall contour, that is, information used to control the position and length of the wall model.

The identification information is information that describes the processing of the graphics in the graphic elements. One identification information includes one or more of instruction mark information, graphic type information, and graphic annotation information. Identification information is usually expressed in a mark and text that people are accustomed to. Such marks and characters are recognized by the computer when they need to be recognized. A set of tokens and text can be used to express one or more messages. For example, when a new graph is marked with only one coordinate, and it has complete and unambiguous coordinate values, it has at least 3 pieces of information, (1) represents the drive mark, and a graphic needs to be added; (2) and represents a graphic type of one Point, (3) The graphic annotation information of this point graphic is that this point is on a specific set of complete coordinates.

In this embodiment, the identification information is used to identify the corresponding graphic element. For example, when the point is marked, the circle may be circled by a circle of no more than 5 mm; when the line is marked, the rectangle may be circled with a width of not more than 5 mm. The modified line, with the break line, intercepts the required part, and needs to drive the partition line to size; when identifying the surface, the surface to be modified is contoured and coiled, and a diagonal line is drawn for explanation. Further, the identification information includes instruction mark information, and the identified graphic element can be processed according to the instruction mark information. The instruction mark information includes one or more of a delete instruction, a modification instruction (including displacement, interception, etc.), an increase instruction, and a query instruction, and the corresponding graphic element is deleted, modified, added, and queried by identifying the instruction. Wait for processing.

The graphic annotation information is an accurate expression of the graphic element, and the labeling style of the graphic annotation information may include a dimension label, an angle label, a coordinate label, an elevation label, an area label, and the like. The dimension is used to indicate the distance between the graphical elements; the angular dimension is used to indicate the angle between the two lines, and the unit information may be degrees or radians; the coordinate label is used to refer to the coordinates of the coordinate points; Annotation is used to identify the elevation; this area is used to mark the area of the face.

The graphical annotation information may include dimension lines and/or dot markers and/or leads and/or text, and the like. In dimensioning and angular dimensioning, dimension lines represent dimensioning styles. The dot mark is used to clearly indicate the point at which the lead is pointed. The lead can point to a clear point, which can be clearly marked with a marker; the lead can also be explicitly pointed to a line. The marked text is located near the dimension line, labeled with the distance or angle represented by the dimension line, or the coordinate value represented by the point marker. The marked text must be a clear number, with or without unit information. The coordinate label must be a set of numbers, and the text of the label can also be a set of calculations.

By setting the definitions of the above points, lines, faces, and identification information, it is possible to identify the composition of the graphic elements and the meanings expressed by the respective parts by identifying the corresponding features, so as to facilitate the generation of subsequent accurate patterns.

After the user finishes drawing the hand drawing with the input device, the steps of recognizing the graphic element step and generating the accurate pattern are entered. The identification of the opponent's drawing, as described above, can identify the entire hand drawing, or can identify the part of the drawing that needs to be revised. It can be understood that the user can also scan the hand-drawn paper-based medium hand drawing and then perform the step of identifying the graphic element.

First, at least one graphic element of the hand drawing is identified; then, the identification information corresponding to the graphic element is identified; and then an accurate pattern corresponding to the graphic element is generated according to the identification information.

Identifying at least one graphical element of the hand drawing, including identifying graphics and identification information in the graphical element, or one of graphics and identification information. Different components of the graphical elements will use different recognition methods.

The identification information corresponding to the graphic element is identified, and the identification information may include instruction mark information, graphic type information, graphic annotation information, and the like.

The graphic type information may be description information of the graphic type in the graphic element, such as: a point, a straight line, an arc, a curve, a surface, or the like, or a combination thereof. It can also be judged according to the labeling style of the graphic annotation information. For example, the coordinate label generally indicates the point, the dimension label generally indicates the line, the radius label or the radians label generally indicates the arc, and the area label generally represents one surface.

When there is no clear graphic type information, it can be considered that the single graphic covering canvas size in the hand drawing is smaller than the minimum range that the user can recognize, that is, the point is determined, and the end points, intersection points, and vertices of the line are also identified as being online. Point. A single graphic overlay canvas is linearly continuous and has a width that is much smaller than the length of the shape and is recognized as a line. The size of the canvas covered by a certain image is larger than the minimum range that the user can recognize, and the coverage is a continuous distribution, which is recognized as a face, and the boundary whose cover is closed is recognized as a boundary line, and the overall content of the cover is filled. Or; a closed line enclosing, can also be identified as a face, the closed line is the boundary line of the face, wherein the enclosed part is the face.

For line graphic elements that do not explicitly label graphic type information, there are many types, such as straight lines, arcs, curves, and so on. For the identification of such a line, the first endpoint, the second endpoint, and at least one sample point between the first endpoint and the second endpoint are identified by the identified graphical element; the first endpoint, the second endpoint, and The sampling point identifies hand-drawn graphic information of the graphic element between the first end point and the second end point, and obtains the graphic type according to the hand-drawn graphic information. Several methods of judgment are provided below, but the actual methods are not limited to the following methods:

The distance relationship between the sampling point and the first end point and the second end point connecting line may be determined by connecting the first end point and the second end point, and hand-drawn graphic information is obtained to determine between the first end point and the second end point. Graphic type; or,

Starting from the first endpoint, continuously calculating the function relationship of the next sampling point until the second endpoint, according to the obtained functional relationship, obtaining hand-drawn graphic information to determine the graphic type between the first endpoint and the second endpoint .

For the recognition of the arc, the sampling point is identified, including the first end point, the second end, the center point, and the like, and the label length is used to identify the arc length, the radius, and the like of the label; or, the geometric method or the tool is used to identify the arc.

For the identification of curves, identify sampling points, such as starting point, ending point, and multiple sampling points. If there is a curve composed of multiple arcs, the segments are identified by arcs. If it is a spline curve, each sample point is smoothly connected in turn, and is identified as a spline curve.

For the identification without a clear line, the first end point and the second end point may be identified and connected, and the length is (x), and a straight line perpendicular to the connecting line is made at the first point. Project the points on the original line on this line. Calculate the length (y) of the two farthest projection points. If the value is less than 0.05, it is recognized as a straight line. If the value is greater than 0.05, take the line point O corresponding to the farthest projection point. The intersection of the line and the plurality of grid lines is taken as a sampling point, and the point set A is composed in the order from the starting point to the O point, wherein there are {a1, a2, a3, ...}. Each projection point of the set A is distanced to the first end point to form a set B of distances, wherein there are {b1, b2, b3, ...}. If the size of each element in the set B is arranged from small to large, the first end point in the judging process is changed to the second end point and judged once again. If the size of each element in the set B is arranged from small to large, it is judged to be an arc, and the point O and the first end point and the second end point are identified as arcs. If the size of each element in the set B is not completely arranged from small to large, it is judged to be a curve. The first endpoint and the second endpoint together with the sample points are identified as a curve.

The recognized type of graphics can also be identified by pre-rendering the graphics on the canvas, because the graphics drawn on the canvas in advance have clear data (it is understandable that explicit data of course contains this graphic type), if the drawn line Referring to the line drawn on the canvas in advance, the line is consistent with the type of the line on which the language is drawn on the canvas.

Graphic annotation information is an accurate representation of graphical elements. Its graphic features are similar to those commonly used by users, including: coordinate labels, dimension labels, angle labels, radius labels, arc length labels, and elevation labels. Its constituent elements include: marking lines, leads, and text. The difference in the constituent elements represents the difference in their type. The difference method is similar to the way people usually distinguish graphic annotations. For example, if the dimension text is a clear angle with a unit, the shape label information must be an angle label. If the label text is a clear label data with a unit, the shape label information must be a Dimensioning, if one end of the dimensioning lead points to a point and the other end points to a line that is almost coincident with the line, it represents a vertical distance from one point to one line. If the dimensioning lead does not coincide or be parallel to the line, it means that the other end points to a point on the line instead of the line, which is the point-to-point distance.

First, judge the annotation form: if the annotation is composed of three intersecting but not closed segments, the segment intersecting the two is the dimension line, the other two are the leads, the intersection is the endpoint of the dimension line, and the dimension is the dimension Label or angular dimension. The dimension line is judged as a line without a clear indication. If the dimension line is a straight line, it is a dimension. If the dimension line is an arc or a curve, it is an angular dimension. If the label is composed of a special elevation symbol and a number, it is marked as an elevation. If the label consists of a point and a lead, and the text is a set of numbers, it is the coordinate label.

Then, the character identifying the information is identified. If the character contains numeric characters (that is, the label data), the size is judged. If the recognized characters contain clear unit information, such as degrees, millimeters, centimeters, meters, etc., it should be the actual size unit information; if there is no unit information in the recognized characters, it is judged whether it is a dimension; if so, the dimension line The two endpoints are initially identified, and the actual distance (a) is calculated according to the scale of the hand drawing and/or the ratio set by the user; the labeled data to be recognized is assumed to be a value after adding a common unit such as millimeters, centimeters, meters, etc. (b) ); the corresponding unit information that determines the value of a and b is the closest, that is, its unit. If the angle is marked, the lead is initially identified, and the angle (a) is calculated according to the scale of the hand drawing and/or the user setting ratio. The number to be recognized, the value after the usual unit such as the addition degree or the radians (b); the unit information corresponding to the value closest to the judgment of a and b is the unit. Calculate the formula. If there is no unknown in the calculation formula, calculate it directly and judge the result directly. If there is an unknown number in the calculation formula, calculate the unknown number first, and then calculate the result size. It can be understood that the size unit information identification method can also be applied to other occasions as an independent technical solution.

Judging the annotation function, if the text in the annotation is a specific number (with or without unit information), the graphic annotation information. If there is no text or the position of the text in the label is a special symbol or code variable, the query is marked.

Identify the lead pointing. Under normal circumstances, when drawing a graphic annotation, the user will deliberately point the lead to the element to be labeled. Therefore, when pointing, the pointing element should be near the pointing end of the lead. If the pointing end points to multiple elements, it should be It is the closest element to the contact point.

Query annotation is the instruction used to query the information in the drawing. The difference between the other methods of marking (modifying, adding, deleting) is that the labeling text is empty or special symbol. The effect is to return the value of the labeled pointing element in the meaning of the label to a specific target (a specific target refers to a variable, or may be a system console, etc.). When the pointing end is an inactive element, the query is performed. If the pointing end is not all inactive elements, the active element is first identified and generated, and when the active element becomes an inactive element, the query is performed.

The order of query annotations and other annotation styles in the actual recognition process is uncertain. However, if a driver annotation contains a query annotation and returns a target variable, the query annotation should be executed first, and the variable value returned by the query annotation is obtained. Identify other labels.

Graphical annotation information is a main method for inputting data elements, and its priority is high. It is convenient and fast to use. It is an innovative technical solution protected by this patent.

Then, according to the identification information, an accurate pattern corresponding to the graphic element is generated: according to the existing element and the graphic type, the effective component group is selected in the component group of the graphic type, and if there are multiple valid component groups, according to the priority The level selects the optimal mode, or the user specifies the constituent feature group, and then the graphic element generates an accurate pattern according to the generation method corresponding to the selected constituent element group.

Determine the reference elements referred to by the graphic annotation information: (1) points and/or lines modified in the canvas without modification instructions; (2) grid lines in the canvas and intersections of the grids; (3) in the canvas Customize the points and/or lines that determine the shape of the position; (4) The graphics that all of the graphical annotation information as the drive end have been driven. The elements of the graphic referred to in (1)(2)(3)(4) above can be used as the elements of the reference end. For the case of (4), it can be understood as: when multiple graphics share a plurality of graphic annotation information together, only one end of the pointing element in which some graphic annotation information is found is an active element, and the other end is an inactive element, then the setting is The inactive end is the reference end, and the active end is the drive end, which first drives the active element pointed to by the annotation information with only one end as the inactive element. After being driven to the position indicated by the annotation, the active element is converted into the inactive element. If both ends of the graphic element are active elements: first drive the other driving label information of the active elements at both ends, and then re-judgment when one end is the inactive end.

It can be understood that some graphic annotation information, such as coordinate annotation, elevation annotation, etc., has only one pointing element in the graphic indication, but it must have a default reference end graphic element. For example, the coordinate label must be marked with the default origin (0, 0) as the reference, and the elevation label is marked with the reference plane's elevation as a reference.

According to the method of obtaining the basic feature group according to the effective component group to generate the accurate pattern, the geometric analysis method of the mathematical operation can be performed by the computer. One of the methods is described below:

The method of analyzing the dimensioning: First, determine the direction of the dimensioning. Generally speaking, if the dimensioning has a line pointing to one end, the direction of the dimensioning will be perpendicular to the direction of the line, and the quantity is dimensioned. Annotate the amount of data and the exact unit composition. Then, the element of the dimensioned reference end uses the mathematical trigonometric function method to calculate the basic feature set of the new point or line obtained by the vector.

Analytic generation method of angle labeling: First, the vector size length (L) of the driving end is unchanged, and the direction angle is composed of the labeled data and the accurate unit (angle O). Then, the constituent element group consisting of L and the angle O and the intersection of the line pointed by the reference end and the line indicated by the driving end (point A), and the other end point of the new vector (B) is obtained according to the mathematical trigonometric function. , the line AB can replace the original drive end line.

Analytic method for elevation labeling: Generally, the elevation label is used, which will involve three-dimensional coordinates. The square can generally be interpreted as perpendicular to the direction of the canvas. It is only necessary to calculate the coordinates of the labeling element of the elevation label perpendicular to the canvas direction. There are coordinate values.

The method of obtaining the basic feature group according to the effective component group to generate the accurate pattern can also be performed by the geometric mapping method, but the general geometric mapping method is flexible and varied. One of the methods is described below:

Dimensioning method of dimensioning: First, the grid and the original image identify the driving line, and the intersection of the lead and the dimension line pointing to the reference end is used as the base point. Keep the base point unchanged, and accurately cut or extend the length of the other end from the base point to the label length. Then, the lead is re-verted to the other end point, and the dimension line is moved parallel to the reference point or the point on the reference line, and the point or line of the drive end is projected onto the new lead. The resulting projection point or line is the point or line that is driven to replace the original line.

The method of drawing the angle mark: First, the mesh identifies the drive line, and finds the intersection (O) of the reference line and the drive line or their extension line, and sets it as a non-mesh point. The cloned element of the reference line is centered on the O point, and the angle is rotated in the direction of the label to obtain a new reference line. Taking O as the origin, measure the distance between each end of the drive line and O, and intercept or extend the original size from the O point to the new reference line. Point the other pointers to the original drive line and rotate them to a new angle. The new reference line that is intercepted or extended replaces the original drive line, and the meta drive line is deleted.

Coordinate labeling: According to the label value, find the actual coordinate point, move the origin to the new position, delete the origin, and the new coordinate point can be out of the canvas.

Elevation Annotation: Sets the plane of the pointing point to the label elevation and recalculates all elevations on the plane of the same pointing point, or all elevations based on this elevation.

Identifying the identification information of a graphical element is an effective way to accurately identify graphical elements and/or data elements: 1 Identifying the elements that are labeled (ie, related to the identifying information) on the inaccurate graphic of the hand-drawn graphical element (point and / Or line); 2 according to the identified elements, find one or more inactive elements as reference elements for the graphic object to be processed; 3 analyze and judge the style and text of the drawn graphic annotation information to determine the need ( 3-1) need to query, or (3-2) need to drive; 4 according to the requirements obtained in the above steps to process the drawings or data: (4-1) need to query to the specified location (variables and / or console and And/or other output; (4-2) constituting the identified element, the reference element, the graphic type information, and the graphic annotation information into a graphic element and/or a data element.

Find the marked features (points and/or lines) on the inaccurate graphic of the hand-drawn graphic elements: Generally, the annotation information will be queried or driven by points and / lines, for example: an existing orthogonal square, The side length is L, and the label now needs to move the square to the right point X to the right and Y to the left point. At this time, we can (method A) mark its lower left vertex (marked feature point) and make relative coordinate labeling (X, Y); (method B) can also mark its upper right corner vertex (labeled element) Point), the relative coordinate labeling (X+L, Y+L); (method C) can also mark the lower side (the edge of the marked element), mark the relative distance with Y, and mark the right side (marked The edge of the feature), and the relative distance is marked X+L. The above three methods (A, B, C) can be completed as required, and different annotation methods correspond to different labeled elements.

Find one or more inactive graphics as reference elements for the objects that need to be processed. In general, any annotation is relative to a fixed (inactive element) reference at this stage. When identifying this annotation, you must find it. This reference.

Analyze and judge the style and text of the drawn graphic annotation information (3-1) Need to query (3-2) Need to drive: Graphic annotation information generally tells the recognition system to do query processing or need to drive processing (drive processing includes, increase , modify), the system will judge according to the label style, whether it is a query label, if the query is marked, the query processing is executed, and if the driver class is marked, the driver processing is executed.

The drawings or data are processed according to the requirements obtained in the above steps: (4-1) If the query is required, the output is to the specified position (variable and/or console and/or other); (4-2) the reference graphic of the labeled information And/or annotated data and the like as graphic elements and/or data elements of the labeled information.

After all the graphic elements are recognized, the recognition ends, and the accurate pattern of the basic feature group data is obtained. According to the preset, the accurate pattern is staked out as a model, and the required drawing form is generated according to the needs. The model can be a two-dimensional model or a three-dimensional model. The drawing can be used to render a 2D or 3D model on a 2D plane using projection, perspective, profiling, animation, etc. This 2D plane is the drawing. Drawings can be output to a variety of output devices (such as: various types of display screens, printers, etc.), and can be added with text descriptions and other information as needed. Of course, if the model being expressed is a two-dimensional model or other case, then the drawing is equivalent to the model, and the step of stakeouting the model from the exact pattern is equivalent to the step of generating the drawing.

It is to be understood that the above-described embodiments are merely illustrative of the preferred embodiments of the present invention, and are not to be construed as limiting the scope of the invention; The above technical features may be freely combined, and various modifications and improvements may be made without departing from the spirit and scope of the invention; therefore, the scope of the claims of the present invention is Equivalent transformations and modifications are intended to be included within the scope of the appended claims.

Claims

A method for generating an accurate pattern based on a hand drawing, comprising the steps of:

S1: identifying at least one graphic element of the hand drawing;

S2: identifying the identification information corresponding to the graphic element;

S3: Generate an accurate pattern corresponding to the graphic element according to the identification information.
A method for generating an accurate pattern according to a hand drawing according to claim 1, wherein said identification information includes instruction flag information; and said step S3 Processing the graphic element according to the instruction mark information.
According to claim 2 The method for generating an accurate pattern according to the hand drawing is characterized in that the instruction mark information includes one or more of a delete instruction, a modification instruction, an increase instruction, and a query instruction.
According to claims 1-3 The method for generating an accurate pattern according to a hand drawing according to any one of the preceding claims, wherein the identification information comprises graphic type information and graphic annotation information;

The step S3 includes:

S3-1: determining whether the graphic type information is included in the identification information, if yes, executing step S3-2, otherwise performing step S3-3 ;

S3-2: determining a graphic type according to the graphic type information, and generating a standard graphic of the graphic of the type;

S3-3: Determine a graphic type corresponding to the graphic element according to the graphic element, and generate a standard graphic of the graphic of the type.
A method for generating an accurate pattern according to a hand drawing according to claim 4, wherein said step S3-3 Includes the following steps:

S3-3-1 : a first endpoint, a second endpoint, and at least one sample point between the first endpoint and the second endpoint in the identified graphical element;

S3-3-2 And identifying, according to the first endpoint, the second endpoint, and the sampling point, hand-drawn graphic information of the graphic element between the first endpoint and the second endpoint, and obtaining a location according to the hand-drawn graphic information The type of graphics.
A method for generating an accurate pattern based on a hand drawing according to claim 5, wherein in said step S3-3-2 Connecting the first end point and the second end point, determining a distance relationship between the sampling point and the first end point and the second end point connecting line, and obtaining the hand-drawn graphic information to determine the first The type of graphics between an endpoint and a second endpoint; or,

Taking the first endpoint as a starting point, continuously calculating a function relationship of the next sampling point until the second endpoint, according to the obtained function relationship, obtaining the hand-drawn graphic information to determine the first end The type of graphics between the point and the second endpoint.
A method for generating an accurate pattern according to a hand drawing according to claim 5, wherein the method further comprises a canvas setting step:

A drawing area is set for the user to draw the hand drawing in the drawing area through an input device.
According to claim 7 The method for generating an accurate pattern according to a hand drawing, characterized in that, by identifying a starting input point of the input device in the drawing area and an input trajectory process, determining that the first end point and the second end are obtained Endpoints and sample points.
The method for generating an accurate pattern according to the hand drawing according to claim 7, wherein the step S3 further comprises:

S3-4: determining whether there is graphic labeling information in the identifier information, if yes, executing step S3-5; otherwise, performing step S3-6 ;

S3-5: Stakeout step S3-2 or S3-3 according to the graphic annotation information Generating the obtained standard pattern to generate the accurate pattern;

S3-6: performing fuzzy calculation according to the position and shape of the brush covering the canvas in the drawing area, and the stakeout step S3-2 or S3-3 The resulting standard pattern is generated to generate the precise pattern.
The method for generating an accurate pattern according to a hand drawing according to claim 9, wherein the step S3-5 Includes the following steps:

S3-5-1: determining whether the graphic annotation information includes unit information, if yes, executing step S3-5-2, otherwise performing step S3-5-3 ;

S3-5-2: generating the accurate pattern according to the annotation data and the unit information in the graphic annotation information;

S3-5-3: according to the annotation data in the graphic annotation information, combined with the proportion of the hand drawing, and/or the user setting ratio, and / Or a common unit, determining the unit information, and generating the accurate pattern according to the label data and the unit information loft.
A method of generating an accurate pattern based on a hand drawing according to claim 1, wherein

In the steps S1 and S2, graphic elements and/or data elements are obtained according to the identified graphic elements;

In the step S3, determining the graphic element and / Or a data element, whether a valid component group is obtained; if a valid component group is obtained, a basic feature group is generated, and the accurate pattern is generated according to the obtained basic feature group data.
A method for generating an accurate pattern according to a hand drawing according to claim 11, wherein in said steps S1, S2 Identifying an element related to the identification information;

Determining a reference element based on the identified element;

Identifying graphic type information and graphic annotation information in the identification information;

The identified element, the reference element, the graphic type information, and the graphic annotation information constitute the graphic element and/or the data element.