CN113989808B - Method and system for selecting specifications of mechanical manufacturing materials based on drawing information processing - Google Patents
Method and system for selecting specifications of mechanical manufacturing materials based on drawing information processing Download PDFInfo
- Publication number
- CN113989808B CN113989808B CN202111243665.1A CN202111243665A CN113989808B CN 113989808 B CN113989808 B CN 113989808B CN 202111243665 A CN202111243665 A CN 202111243665A CN 113989808 B CN113989808 B CN 113989808B
- Authority
- CN
- China
- Prior art keywords
- product
- line
- image
- preset
- lines
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for selecting specifications of a mechanical manufacturing material based on drawing information processing. The method comprises the following steps: converting a drawing to be processed into a drawing image in a preset picture format; performing image-text separation pretreatment on a drawing image based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in the drawing image, and acquiring character data and image coordinate data; comprehensively analyzing the character data and the image data, and judging whether the drawing image has size loss: if not, selecting the material specification; and if so, jumping to a preset imaging DFM module, and marking the position of size loss in the drawing image. The scheme provided by the invention can improve the efficiency and accuracy of drawing analysis and material specification selection, reduce the error rate caused by human factors and further improve the manufacturing efficiency of products.
Description
Technical Field
The invention relates to the field of machine manufacturing, in particular to a method and a system for selecting specifications of a machine manufacturing material based on drawing information processing.
Background
DFM (Design for manufacturing) means a Design-oriented Design, i.e., a Design for manufacturability analysis, that starts with improving the manufacturability of a part, making the part and various processes easy to manufacture, low in manufacturing cost, efficient, and low in cost rate. DFM means that the product design needs to meet the requirements of product manufacturing with good manufacturability so that the product is manufactured with the lowest cost, the shortest time, and the highest quality. According to different manufacturing processes of products, the design for manufacturing can be divided into a design for injection molding, a design for stamping, a design for die casting, and the like.
DFM is the core technology of parallel engineering nowadays, because design and manufacture are the two most important links in the life cycle of a product, and the parallel engineering is to consider factors such as manufacturability and assembly of the product when starting design. DFM is the most important support tool in parallel engineering. The key of the method is the process analysis of design information, the evaluation of manufacturing rationality and the proposal of improving the design.
The traditional machining product cost calculation needs complicated and complicated procedures such as drawing evaluation, material sample selection, manufacturability design analysis, process establishment, cost calculation and the like by depending on enterprise technicians. The processing evaluation and the cost calculation are carried out manually, so that the problem of calculation error is easy to occur, and the problems of high evaluation error rate, low evaluation efficiency and the like are caused. In addition, businesses often need to spend a great deal of time and money retraining new technicians for evaluation each time an associated technician leaves their job.
And the drawing evaluation and the material sample selection are indispensable steps before the analysis of the manufacturability design. And the material sample can be processed into a product for manufacturability design analysis only by extracting product information and processing requirements in the drawing. Some automatic quotation software appears in the prior art, but the automatic quotation software can only assist enterprises to solve the cost calculation part under a preset scheme. Meanwhile, such software also needs a skilled worker with abundant experience to perform works such as drawing evaluation and material selection in advance, and the problems cannot be completely solved.
Therefore, a solution to replace manual drawing analysis and material selection is urgently needed to solve the above problems.
Disclosure of Invention
In view of this, the invention provides a method and a system for selecting specifications of a machine manufacturing material based on drawing information processing, and the specific scheme is as follows:
a mechanical manufacturing material specification selection method based on drawing information processing comprises the following steps:
obtaining a drawing to be processed of a product, and converting the drawing to be processed into a drawing image in a preset picture format;
performing image-text separation pretreatment on the drawing image based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in the drawing image, and acquiring first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system;
comprehensively analyzing the characters and the first coordinate data, the images and the second coordinate data, and judging whether the drawing image has size loss:
if not, selecting the material specification; if so, jumping to a preset imaging DFM module, and marking the position of size loss in the drawing image;
the material specification selection comprises: acquiring the appearance characteristics of a product, calculating the volume of the product, and selecting a material sample meeting preset conditions from a preset material database as a processing sample according to the appearance characteristics and the volume of the product; the material database can acquire the information of the material samples which can be purchased in the current market, and screen and store the material samples with various specifications.
In a specific embodiment, the material specification selection specifically includes:
constructing a 3D model of the product according to the drawing image, acquiring the appearance characteristics of the product, and calculating the product volume of the product;
selecting all material samples with the outer dimension larger than or equal to the outer dimension of the product from a preset material database based on the appearance characteristics and the product volume, and calculating the material cost and the cutting volume of each material sample, wherein the cutting volume is the volume required to be cut off by the material samples to obtain the product;
and comprehensively analyzing the material cost, the processing step and the processing cost required by each material sample based on the material cost and the cutting volume, and selecting the most cost-effective material sample as a processing sample.
In a specific embodiment, the "converting the drawing to be processed into a drawing image in a preset picture format" specifically includes:
if the drawing to be processed is an electronic version drawing, directly converting the drawing to be processed into an image file format;
and if the drawing to be processed is a paper version drawing, performing image scanning on the drawing to be processed and storing the drawing to be processed into an image archive format.
In one embodiment, "determining whether the drawing image has a size deficiency" includes:
judging whether each product line has a preset corresponding relation with a certain size line or not based on the characters, the first coordinate data, the images and the second coordinate data:
if yes, the size line represents the size of the current product line; if not, the current product line lacks the size;
identifying and judging whether each dimension line has dimension marks:
if not, the current dimension line lacks the dimension.
In a specific embodiment, the characters include materials, product dimensions, dimensional boundaries and lines, geometric signatures, references, price characteristics, surface finish requirements, shape and positional tolerance range values;
the image includes only the product line of the product.
In a particular embodiment, identifying the product line includes:
based on the fact that the product line is thicker than the non-product line and the line of the product line is a closed solid line, non-arrow solid lines with thickness degrees obviously not meeting preset thickness conditions are eliminated, and then in the remaining non-arrow solid lines, whether head and tail coordinates of each non-arrow solid line intersect with other non-arrow solid lines or not is judged:
if so, the line is a product line; if not, the line is a non-product line.
In one embodiment, the recognition algorithm comprises:
dividing lines on a drawing into basic lines and difficult lines, wherein the basic lines comprise lines with obvious characteristics, and the difficult lines comprise lines which are difficult to judge except the basic lines;
identifying the base line based on characteristics of the product line, the dimension line, and the dimension boundary line;
and identifying the difficult line based on the trained identification algorithm.
In a specific embodiment, the training process of the recognition algorithm specifically includes:
acquiring drawings of different marking methods, manually marking characters and images, and acquiring a drawing training sample;
and carrying out supervised training on an initial machine learning algorithm through the drawing training sample to obtain the recognition algorithm.
In a specific embodiment, the character is selected through a preset box, and the first coordinate data is the coordinate of the box where the character is located;
and setting the area of the square box according to the coverage area of the character.
A mechanical manufacturing material specification selection system based on drawing information processing comprises,
drawing information acquisition module: the system comprises a processing unit, a processing unit and a processing unit, wherein the processing unit is used for acquiring a to-be-processed drawing of a product and converting the to-be-processed drawing into a drawing image with a preset picture format;
performing image-text separation pretreatment based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in the drawing image, and acquiring first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system;
a drawing analysis module: the DFM module is arranged and used for comprehensively analyzing the characters, the first coordinate data, the images and the second coordinate data and judging whether the drawing image has size loss or not:
if not, selecting the material specification; if yes, jumping to the DFM module, and marking the position of the missing size in the drawing image;
a material specification selection module: the device is used for obtaining the appearance characteristics of a product, calculating the volume of the product, selecting a material sample meeting preset conditions from a preset material database as a processing sample according to the appearance characteristics and the volume of the product, wherein the preset material database can obtain the information of the material sample which can be purchased in the current market, and screening and storing various material samples.
Has the advantages that:
the invention provides a method and a system for selecting specifications of a mechanical manufacturing material based on drawing information processing, which replace manual identification by an algorithm and realize intellectualization of drawing analysis and material sample selection. The material specification selection module can automatically select all material samples meeting the specification requirements, and selects the most cost-effective scheme from the material samples, manual calculation and selection are not needed, and the intellectualization of material specification selection is realized. The production method has the advantages that drawing information acquisition and drawing analysis are carried out on the product drawing, intelligent processing of the drawing is achieved, processing assessment and cost calculation are achieved through an algorithm, manual participation is not needed, and the production scheme of the product can be obtained. The scheme provided by the invention can improve the efficiency and accuracy of drawing analysis and material specification selection, reduce the error rate caused by human factors and further improve the manufacturing efficiency of products.
Drawings
FIG. 1 is a flow chart illustrating a method for selecting a specification of a machine-made material according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a drawing information extraction section and a drawing analysis section according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a material specification selection section according to an embodiment of the present invention;
FIG. 4 is an illustration of a drawing provided by an embodiment of the present invention;
FIG. 5 is an exemplary diagram of a drawing after pre-processing of image-text separation according to an embodiment of the present invention;
FIG. 6 is an exemplary diagram of a DFM drawing for size deficiency according to an embodiment of the present invention;
FIG. 7 is a block diagram of a selected specification for a machine-made material according to an embodiment of the present invention.
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Reference numerals: 1-a drawing information acquisition module and 2-a drawing analysis module; and 3-selecting a material specification module.
Detailed Description
Hereinafter, various embodiments of the present disclosure will be described more fully. The present disclosure is capable of various embodiments and of being practiced with modification and alteration. However, it should be understood that: there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but rather, the disclosure is to cover all modifications, equivalents, and/or alternatives falling within the spirit and scope of the various embodiments of the present disclosure.
The terminology used in the various embodiments of the disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the various embodiments of the disclosure. As used herein, the singular forms are intended to include the plural forms as well, unless the context clearly indicates otherwise. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of the present disclosure belong. The terms (such as those defined in commonly used dictionaries) should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein in various embodiments.
Example 1
The embodiment 1 of the invention provides a method for selecting specifications of a mechanical manufacturing material based on drawing information processing, which replaces manual identification with an algorithm to realize drawing analysis intellectualization and material sample selection intellectualization, and is specifically shown in the accompanying drawings 1-6 of the specification. The flow schematic diagram of the method is shown in the attached figure 1 of the specification, and the specific scheme is as follows:
101. acquiring a to-be-processed drawing of a product, and converting the to-be-processed drawing into a drawing image in a preset picture format;
102. performing image-text separation pretreatment on a drawing image based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in the drawing image, and acquiring first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system;
103. comprehensively analyzing the characters and the first coordinate data, the images and the second coordinate data, and judging whether the drawing to be processed has size loss:
104. if so, jumping to a preset imaging DFM module, and marking the position of size loss in the drawing image;
105. if not, selecting the material specification; material specification selection includes: the method comprises the steps of obtaining the appearance characteristics of a product, calculating the volume of the product, and selecting a material sample meeting preset conditions from a preset material database as a processing sample according to the appearance characteristics and the volume of the product, wherein the preset material database stores a plurality of material samples.
The method for selecting the specification of the machine manufacturing material based on drawing information processing mainly comprises three parts, namely drawing information acquisition, drawing analysis and material specification selection. Each part can be realized through an algorithm preset by the system without manual identification. The flow chart of the drawing information acquisition and drawing analysis part is shown in the specification and the attached figure 2, and the flow chart of the material specification selection part is shown in the specification and the attached figure 3.
Specifically, step 101 obtains a drawing to be processed, and performs format conversion on the drawing to be processed. The drawing to be processed comprises electronic layout paper and paper layout paper. And if the drawing to be processed is the electronic version drawing, directly converting the drawing to be processed into an image file format. And if the to-be-processed drawing is a paper version drawing, performing image scanning on the to-be-processed drawing and storing the to-be-processed drawing in an image archive format.
In this embodiment, the drawing is converted into a drawing image, and the predetermined format is an image file format, such as a BMP image file format.
Specifically, in step 102, image-text separation preprocessing is performed on the drawing image through a preset recognition algorithm, a coordinate system is constructed, characters and images in the drawing image are recognized, and first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system are obtained.
The drawing of fig. 4 in the specification is taken as an example drawing. After the image-text separation pretreatment, the drawing is processed into a schematic diagram shown in the attached figure 5 in the specification. In the specification, fig. 4 can be seen as providing a rectangular part with a circular hollow-out middle part, and the dimension mark of the rightmost dimension line in the drawing is deleted.
In the present embodiment, information on the drawing sheet is divided into characters and images. The image includes only the product line of the product, i.e. the line describing the shape of the product. The characters include materials, product dimensions, dimension boundary lines and dimension lines, geometric feature symbols, reference standards, price characteristics, surface processing requirements, shape and position tolerance range values, and the like, and information other than the product lines can be classified as characters. The characters represent information of the product, and the images represent a shape of the product. Preferably, the characters and the first coordinate data can be stored in a preset character database, and the images and the second coordinate data can be stored in an image database, so that the drawing analysis module can conveniently acquire drawing information. For example, the character and the first coordinate data may be stored in a txt format in a character repository, and the image and the second coordinate data may be stored in a dwg format in an image repository. In the specification of fig. 4, the product lines are four thick straight lines of a square structure and thick round lines of a round structure, and the product size, the size boundary line and the size line are all characters except for the product size, the size boundary line and the size line. The coordinate system is constructed as shown in the attached figure 5, and the size of the coordinate system can be consistent with the size of the product or can be inconsistent with the size of the product. For example, the length of the product line on the drawing sheet is 4.65mm, and the head and tail coordinate data of the product line in the coordinate system of fig. 5 are (1, 1) and (5.65, 1), whereas the head and tail coordinates of the product line in other coordinate systems may be (1, 2) and (11.3, 2). Namely, the coordinate system can have a single set of unit standard, so as to avoid being confused with the size information of the product.
The character is selected through a preset square frame, and the first coordinate data is the coordinate of the square frame where the character is located. Since characters include various data formats, such as data, lines, etc., different characters have different sizes and corresponding boxes are different. And setting the area of the square box by identifying the size of the character so that the square box completely selects the character. For example, product dimensions are often in the form of numeric values, dimension lines are arrowed, boxes of dimension lines need to completely cover dimension lines, and boxes of product dimensions need to completely cover product dimensions. And (3) learning and identifying various characters in the drawing by using various algorithms, and screening and classifying the characters. As shown in fig. 5, in the constructed coordinate system, each character and image on the drawing has uniquely determined coordinate data. The preset box selects characters, each character having fixed coordinate data. For example, the character c1 is framed by a box, the coordinate data of the four corners of the box are fixed, and the position of the character c1 can be determined by the coordinate data of the box. Similarly, the dimension lines and dimension boundaries are also enclosed by boxes, as shown in FIG. 5 of the specification.
Lines on the drawings include product lines, dimension boundary lines, dimension lines, and other lines. The dimension lines are arrow lines, including single arrow lines and double arrow lines. The dimensional boundaries generally match dimensional lines, with arrowed lines connecting the dimensional boundaries. Other lines include boxes labeled with specific parameters such as roughness. Each product line needs to be marked with a size, and can be precisely manufactured finally. For example, a product line in a straight line shape may represent a relevant parameter by a fit of a size mark, a size boundary line, and a size line; the arc-shaped product line can represent related parameters through the position of the circle center, the included angle and the like. In this embodiment, each product line is identified by the variation of the line and the dimensioning. For example, a straight line and an arc intersect to form a continuous line, the length of the straight line and relevant parameters of the arc are marked by a standard drawing marking method, at the moment, two size marks can be identified through an identification algorithm, and the line is judged to be formed by two product lines.
Based on the fact that the product lines are all closed solid lines and are thicker than non-product lines, the identification of the product lines comprises the following steps: firstly, excluding the non-arrow solid lines with thickness degrees obviously not meeting the preset thickness condition, and then judging whether the head and tail coordinates of each non-arrow solid line are intersected with other lines in the rest non-arrow solid lines: if so, the line is a product line; if not, the line is a non-product line. The line thickness degree of the product line is obviously thicker than the solid lines such as the dimension boundary lines, therefore, a preset thickness condition is set according to the thickness of the product line, non-arrow solid lines which do not meet the preset thickness condition are firstly excluded, and the solid lines such as the dimension boundary lines are excluded. Then, in the remaining non-arrow solid lines, it is determined whether or not the non-arrow solid lines intersect with each other. It should be noted that the excluded solid line without an arrow is not within the judgment range where the subsequent solid lines intersect. In the drawings, a closed product line can form a complete product. Based on the principle, the embodiment constructs a coordinate system when performing the image-text separation preprocessing. Each character, each line on the drawing sheet, has unique and definite coordinates on the coordinate system. The coordinate system is arranged so that even if the character is separated from the image, the unique and definite corresponding relation can be kept.
Each product line will have an intersection with other product lines, and it appears in the coordinate system that the head and tail coordinates of each product line will intersect with other product lines. In this embodiment, the head and tail coordinates of the product line are the start point coordinates and the end point coordinates of the product line, that is, the coordinates of the two end points of the product line. The standard dimension lines are all arrow lines, excluding arrow lines, and product lines are looked up in the non-arrow solid lines. By judging whether the head and tail coordinates of each non-arrow solid line intersect with other lines: if yes, the line is a product line; if not, the line is a non-product line.
In this embodiment, the recognition algorithm performs image-text separation preprocessing on the drawing image by means of combining feature recognition and sample learning, and recognizes, screens and classifies characters and images. Specifically, lines in the image of the drawing are divided into base lines and difficult lines. The basic line includes a line whose features are obvious and which can be identified by the features. For example, the base line may include a dimension line, a portion of a product line, and a dimension demarcation line. The size line is provided with an arrow, the size boundary line and the arrow have intersection points, the size line and the size boundary line can be identified through the characteristic of the arrow, intersection points exist between the head and tail coordinates of the product line and other product lines, the single line is judged by classification, and then the product line is judged by identification.
However, different drawings are drawn by different engineers, and different drawing styles result in different labeling methods. For example, in some drawings, some special product parameters may be enclosed by a closed line, such as a circle, and it is difficult to determine whether the line is a product line through the feature recognition of the basic line. Therefore, in order to adapt to different labeling methods, the recognition algorithm of the embodiment has sample learning capacity, and supervised training is performed on the recognition algorithm through a large number of drawing samples of different labeling methods.
Specifically, the training process includes: acquiring drawings of different labeling methods, manually labeling characters and images, and acquiring a drawing training sample; and carrying out supervised training on the initial machine learning algorithm through a drawing training sample to obtain a trained recognition algorithm. The recognition algorithm of the present embodiment includes, but is not limited to, any intelligent algorithm with sample learning capability.
Specifically, 103, comprehensively analyzing the characters and the first coordinate data, and the images and the second coordinate data, and judging whether the drawing to be processed has size loss: if so, jumping to a preset imaging DFM module, and marking the position of size loss in the drawing image; if not, selecting the material specification.
The character and the first coordinate data are obtained from a character database, and the image and the second coordinate data are obtained from an image database. And analyzing the character and the first coordinate data, the image and the second coordinate data, and judging whether each product line has a preset corresponding relation with a certain dimension line and whether the dimension line has missing dimension marks or not, thereby judging whether the dimension is missing in the drawing image or not.
In the manufacturing field, each product line on the drawing needs to be marked with a dimension, and the dimension line is positioned near the product line and has a corresponding relation with the product line. Taking a common straight line as an example, each product line in a straight line needs to be marked with the length of the straight line, a dimension boundary line extends outwards from the start point coordinate and the end point coordinate of the product line, the dimension line is between the two dimension boundary lines, and the dimension mark is located near the dimension line to indicate the length of the product line. Based on the characteristic, the head and tail coordinates of the dimension line can be searched through the characters and the first coordinate data, and the head and tail coordinates of the product line can be searched through the image and the second coordinate data; judging whether the head and tail coordinates of each product line have a preset corresponding relation with the head and tail coordinates of a certain size line, if so, indicating the size of the current product line by the size line; and if not, the current product line lacks the size. Because the product line and the dimension line are in parallel relation and have equal length, the head and tail coordinates of the product line and the head and tail coordinates of the dimension line have corresponding relation. If the product line is a horizontal straight line, the start point coordinate of the dimension line is the same as the start point coordinate of the product line, and the end point coordinate of the dimension line is the same as the longitudinal coordinate of the end point coordinate of the product line, and the length of the product line calculated according to the coordinates is equal to the length of the dimension line. Based on the characteristics, the product line corresponding to the size line can be judged. By identifying and judging whether a dimension mark exists on each dimension line: if not, the current dimension line lacks the dimension.
If the size is missing, jumping to a preset imaging DFM module to perform corresponding intelligent processing, labeling the drawing image, and feeding the result back to an engineer for subsequent processing. The DFM module will give relevant recommendations to compensate for the missing size. Specifically, in the specification, fig. 4, the rightmost dimension line lacks dimension marks, that is, the drawing has dimension loss. After the missing size analysis is performed in step 103, labeling is performed at the missing size. The drawing after labeling is schematically shown in the attached figure 6 of the specification, a dimension line with a missing dimension label is marked at the position marked in the drawing, and a DFM is marked beside the dimension line: missing size ", and framing the DFM suggestion with a square frame.
If there is no size loss, material specification selection is performed. The material specification selection specifically comprises:
and constructing a 3D model of the product according to the drawing image, acquiring the appearance characteristics of the product, and calculating the product volume. CAD software can be used to convert two-dimensional images in drawings to three-dimensional models. And 3D models are built, so that the appearance characteristics of the product can be visually obtained, the volume of the product can be conveniently calculated, and the parameters of all parts of the product can be known.
And selecting a processing sample from a preset material database based on the appearance characteristics and the product volume, wherein the processing sample is a material sample with the outer dimension larger than or equal to the outer dimension of the product. And calculating the material cost and the cut volume of each material sample, wherein the cut volume is the volume to be cut off for obtaining the product from the material sample. In this embodiment, the material database can acquire information of material samples that can be purchased in the current market, and screen and store a plurality of material samples of different specifications. For example, the corresponding information can be crawled from a predetermined website, which can record the material samples currently on the market, so that the material samples in the material database can be purchased from the market. Different specifications of material sample materials may differ in sample material, sample volume, material cost, sample processing method, and the like. For example, a hexagonal cylinder part with H side to side and length l needs to be machined. Firstly, to remove the material sample with the outside dimension not meeting the requirement, the outside dimension of the material sample is larger than or equal to and closest to the outside dimension of the hexagonal cylinder part, and all materials with proper specifications are selected. Secondly, for the material samples meeting the size requirement, the volume of the hexagonal cylinder part required to be cut obtained by the material samples is calculated, and the cutting volumes of different material samples can be different.
And comprehensively analyzing the material cost, the processing step and the processing cost required by each material sample based on the material cost and the cutting volume, and selecting the material sample with the most cost-benefit. Different material samples may need to be processed by different processing methods by different processing equipment, and the processing cost may be different. Therefore, it is necessary to select the most cost-effective material sample by considering various factors, combining the feasibility of the material sample and the processing capability of the equipment, and analyzing the material cost, the processing step, and the processing cost required for each material sample. In addition, the production quantity of the product needs to be considered, such as single-piece production and mass production, and even if the single-piece processing cost is high by adopting a certain method, the method can be adopted for mass production.
The embodiment provides a mechanical manufacturing material specification selection method based on drawing information processing, and the intellectualization of drawing analysis and material sample selection is realized by replacing manual identification with an algorithm. The material specification selection module can automatically select all material samples meeting the specification requirements, selects the most cost-effective scheme from the material samples, does not need manual calculation and selection, and realizes the intellectualization of material specification selection. The production method has the advantages that drawing information acquisition and drawing analysis are carried out on the product drawing, intelligent processing of the drawing is achieved, processing assessment and cost calculation are achieved through an algorithm, manual participation is not needed, and the production scheme of the product can be obtained. The scheme provided by the embodiment can improve the efficiency and accuracy of drawing analysis and material specification selection, reduce the error rate caused by human factors, and further improve the manufacturing efficiency of products.
Example 2
The embodiment 2 of the invention discloses a mechanical manufacturing material specification selection system based on drawing information processing, and adopts the mechanical manufacturing material specification selection method based on drawing information processing provided by the embodiment 1. On the basis of the embodiment 1, the method of the embodiment 1 is systematized, the structural module is as shown in the attached figure 7 of the specification, and the specific scheme is as follows:
a mechanical manufacturing material specification selection system based on drawing information processing comprises a drawing information acquisition module 1, a drawing analysis module 2 and a material specification selection module 3.
Drawing information acquisition module 1: the system comprises a processing unit, a processing unit and a processing unit, wherein the processing unit is used for acquiring a to-be-processed drawing of a product and converting the to-be-processed drawing into a drawing image in a preset picture format;
and performing image-text separation pretreatment based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in the drawing image, and acquiring first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system.
Drawing analysis module 2: the system is provided with a DFM module for comprehensively analyzing the characters and the first coordinate data, the images and the second coordinate data and judging whether the drawing to be processed has size loss:
if not, selecting the material specification; if so, jumping to a DFM module, and marking the position of the missing size in the drawing image;
material specification selection module 3: the method is used for obtaining the appearance characteristics of the product, calculating the volume of the product, and selecting the material samples meeting preset conditions from a preset material database according to the appearance characteristics and the volume of the product, wherein the preset material database stores the material samples with various specifications. In this embodiment, the material database can acquire information of material samples available in the current market, and screen and store a plurality of material samples with different specifications. For example, the corresponding information can be crawled from a predetermined website, which can record the material samples currently on the market, so that the material samples in the material database can be purchased from the market.
The embodiment provides a mechanical manufacturing material specification selection system based on drawing information processing, and adopts the mechanical manufacturing material specification selection system based on drawing information processing provided by the embodiment 1. On the basis of the embodiment 1, the method of the embodiment 1 is systematized, so that the method has more practical applicability.
The invention provides a method and a system for selecting specifications of a mechanical manufacturing material based on drawing information processing, which replace manual identification by an algorithm and realize intellectualization of drawing analysis and material sample selection. The material specification selection module can automatically select all material samples meeting the specification requirements, and selects the most cost-effective scheme from the material samples, manual calculation and selection are not needed, and the intellectualization of material specification selection is realized. The production method has the advantages that drawing information acquisition and drawing analysis are carried out on the product drawing, intelligent processing of the drawing is achieved, processing assessment and cost calculation are achieved through an algorithm, manual participation is not needed, and the production scheme of the product can be obtained. The scheme provided by the invention can improve the efficiency and accuracy of drawing analysis and material specification selection, reduce the error rate caused by human factors and further improve the manufacturing efficiency of products.
Those skilled in the art will appreciate that the drawings are merely schematic representations of preferred embodiments and that the blocks or flowchart illustrations are not necessary to practice the present invention. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario may be distributed in the devices in the implementation scenario according to the description of the implementation scenario, or may be located in one or more devices different from the present implementation scenario with corresponding changes. The modules of the implementation scenario may be combined into one module, or may be further split into multiple sub-modules. The above-mentioned invention numbers are merely for description and do not represent the merits of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of the present invention, however, the present invention is not limited thereto, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present invention.
Claims (9)
1. A mechanical manufacturing material specification selection method based on drawing information processing is characterized by comprising the following steps:
obtaining a to-be-processed drawing of a product, and converting the to-be-processed drawing into a drawing image in a preset picture format;
performing image-text separation pretreatment on the drawing image based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in the drawing image, and acquiring first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system;
comprehensively analyzing the characters and the first coordinate data, the images and the second coordinate data, and judging whether the drawing image has size loss:
if not, selecting the material specification; if so, jumping to a preset imaging DFM module, and marking the position of size loss in the drawing image;
the material specification selection comprises: acquiring the appearance characteristics of a product, calculating the volume of the product, and selecting a material sample meeting preset conditions from a preset material database as a processing sample according to the appearance characteristics and the volume of the product; the material database can acquire the information of the material samples which can be purchased in the current market, and screen and store the material samples with various specifications;
wherein the characters include materials, product dimensions, dimensional boundaries and lines, geometric signatures, references, price features, surface finish requirements, shape and position tolerance range values; the image includes only the product line of the product.
2. The method of selecting a machine-building material specification according to claim 1, wherein the material specification selection specifically comprises:
constructing a 3D model of the product according to the drawing image, acquiring the appearance characteristics of the product, and calculating the product volume of the product;
selecting all material samples with the outer dimension larger than or equal to the outer dimension of the product from a preset material database based on the appearance characteristics and the product volume, and calculating the material cost and the cutting volume of each material sample, wherein the cutting volume is the volume required to be cut off by the material samples to obtain the product;
and comprehensively analyzing the material cost, the processing step and the processing cost required by each material sample based on the material cost and the cutting volume, and selecting the material sample with the most cost-effectiveness as a processing sample.
3. The method for selecting specifications of a machine-building material according to claim 1, wherein converting the drawing to be processed into a drawing image in a predetermined picture format specifically comprises:
if the drawing to be processed is an electronic version drawing, directly converting the drawing to be processed into an image file format;
and if the drawing to be processed is a paper version drawing, performing image scanning on the drawing to be processed and storing the drawing to be processed into an image archive format.
4. The method of claim 1, wherein the step of determining whether the drawing image has a dimensional deficiency includes:
judging whether each product line has a preset corresponding relation with a certain size line or not based on the characters, the first coordinate data, the images and the second coordinate data:
if so, the size line represents the size of the current product line; if not, the current product line lacks the size;
identifying and judging whether each dimension line has dimension marks:
if not, the current dimension line lacks the dimension.
5. The machine-made material specification selection method of claim 1, wherein identifying the product line comprises:
based on the fact that the product line is thicker than the non-product line and the product line is a closed solid line, the non-arrow solid lines with the thickness degree obviously not meeting the preset thickness condition are eliminated, and then in the remaining non-arrow solid lines, whether the head and tail coordinates of each non-arrow solid line are respectively intersected with other non-arrow solid lines is judged:
if so, the line is a product line; if not, the line is a non-product line.
6. The machine-made material specification selection method of claim 5, wherein the identification algorithm comprises:
dividing lines on a drawing into basic lines and difficult lines, wherein the basic lines comprise lines with obvious characteristics, and the difficult lines comprise lines which are difficult to judge except the basic lines;
identifying the base line based on characteristics of the product line, the dimension line, and the dimension boundary line;
and identifying the difficult line based on the trained identification algorithm.
7. The method of claim 6, wherein the training of the recognition algorithm specifically comprises:
acquiring drawings of different labeling methods, manually labeling characters and images, and acquiring a drawing training sample;
and carrying out supervised training on an initial machine learning algorithm through the drawing training samples to obtain the recognition algorithm.
8. The machine-building material specification selection method according to claim 1, wherein the character is selected by a preset box, and the first coordinate data is coordinates of the box where the character is located;
and setting the area of the square box according to the coverage area of the character.
9. A mechanical manufacturing material specification selection system based on drawing information processing is characterized by comprising,
drawing information acquisition module: the system comprises a processing unit, a processing unit and a processing unit, wherein the processing unit is used for acquiring a to-be-processed drawing of a product and converting the to-be-processed drawing into a drawing image with a preset picture format;
performing image-text separation pretreatment based on a preset identification algorithm, constructing a coordinate system, identifying characters and images in a drawing image, and acquiring first coordinate data of the characters in the coordinate system and second coordinate data of the images in the coordinate system;
a drawing analysis module: the DFM module is arranged and used for comprehensively analyzing the characters, the first coordinate data, the images and the second coordinate data and judging whether the drawing images have size loss or not:
if not, selecting the material specification; if yes, jumping to the DFM module, and marking the position of the missing size in the drawing image;
a material specification selection module: the system comprises a product processing module, a product database and a data processing module, wherein the product processing module is used for acquiring the appearance characteristics of a product, calculating the volume of the product, and selecting a material sample meeting preset conditions from a preset material database as a processing sample according to the appearance characteristics and the volume of the product, and the preset material database can acquire the information of the material sample which can be purchased in the current market, and screen and store a plurality of material samples;
wherein the characters comprise material, product dimensions, dimensional boundaries and lines, geometric signatures, references, price features, surface finish requirements, shape and positional tolerance range values; the image includes only the product line of the product.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111243665.1A CN113989808B (en) | 2021-10-25 | 2021-10-25 | Method and system for selecting specifications of mechanical manufacturing materials based on drawing information processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111243665.1A CN113989808B (en) | 2021-10-25 | 2021-10-25 | Method and system for selecting specifications of mechanical manufacturing materials based on drawing information processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113989808A CN113989808A (en) | 2022-01-28 |
| CN113989808B true CN113989808B (en) | 2022-11-18 |
Family
ID=79741232
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111243665.1A Active CN113989808B (en) | 2021-10-25 | 2021-10-25 | Method and system for selecting specifications of mechanical manufacturing materials based on drawing information processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113989808B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114952059B (en) * | 2022-06-08 | 2023-05-23 | 深圳市大族机器人有限公司 | Automatic welding method and system based on cooperative robot |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105302931A (en) * | 2014-06-30 | 2016-02-03 | 上海神机软件有限公司 | Recognition system and method for construction engineering drawing, and template arrangement system and method |
| CN109117713A (en) * | 2018-06-27 | 2019-01-01 | 淮阴工学院 | A kind of drawing printed page analysis of full convolutional neural networks and character recognition method |
| CN110969111A (en) * | 2019-11-28 | 2020-04-07 | 苏州安永数据科技有限公司 | Automatic identification and classification method for mechanical part digital drawing |
| CN111275501A (en) * | 2020-03-06 | 2020-06-12 | 刘共清 | Intelligent valuation method based on building scheme |
| CN112486383A (en) * | 2020-11-26 | 2021-03-12 | 万翼科技有限公司 | Picture examination sharing method and related device |
| CN112486384A (en) * | 2020-11-27 | 2021-03-12 | 万翼科技有限公司 | Picture examination processing method and related device |
| CN112580316A (en) * | 2020-12-24 | 2021-03-30 | 苏州特文思达科技有限公司 | Computer-aided steel grating product information intelligent identification and statistics technology |
| CN113128457A (en) * | 2021-04-30 | 2021-07-16 | 杭州品茗安控信息技术股份有限公司 | Building model identification method, system and related device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9367063B2 (en) * | 2013-10-17 | 2016-06-14 | Plethora Corporation | Method for implementing design-for-manufacturability checks |
| JP6329424B2 (en) * | 2014-05-01 | 2018-05-23 | ミサワホーム株式会社 | Building specification confirmation support system |
| CN105975562A (en) * | 2016-05-03 | 2016-09-28 | 水木智博（北京）网络信息科技有限公司 | Method and apparatus for automatically generating budget table of engineering drawing |
| CN108748400A (en) * | 2018-05-05 | 2018-11-06 | 佛山市德法科技有限公司 | A kind of full-automatic water cutting production equipment and its processing method |
| US11783389B2 (en) * | 2019-12-06 | 2023-10-10 | Proto Labs, Inc. | Methods and systems for predicting a price of any subtractively manufactured part utilizing artificial intelligence at a computing device |
-
2021
- 2021-10-25 CN CN202111243665.1A patent/CN113989808B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105302931A (en) * | 2014-06-30 | 2016-02-03 | 上海神机软件有限公司 | Recognition system and method for construction engineering drawing, and template arrangement system and method |
| CN109117713A (en) * | 2018-06-27 | 2019-01-01 | 淮阴工学院 | A kind of drawing printed page analysis of full convolutional neural networks and character recognition method |
| CN110969111A (en) * | 2019-11-28 | 2020-04-07 | 苏州安永数据科技有限公司 | Automatic identification and classification method for mechanical part digital drawing |
| CN111275501A (en) * | 2020-03-06 | 2020-06-12 | 刘共清 | Intelligent valuation method based on building scheme |
| CN112486383A (en) * | 2020-11-26 | 2021-03-12 | 万翼科技有限公司 | Picture examination sharing method and related device |
| CN112486384A (en) * | 2020-11-27 | 2021-03-12 | 万翼科技有限公司 | Picture examination processing method and related device |
| CN112580316A (en) * | 2020-12-24 | 2021-03-30 | 苏州特文思达科技有限公司 | Computer-aided steel grating product information intelligent identification and statistics technology |
| CN113128457A (en) * | 2021-04-30 | 2021-07-16 | 杭州品茗安控信息技术股份有限公司 | Building model identification method, system and related device |
Non-Patent Citations (1)
| Title |
|---|
| 基于快速设计的注塑模报价***的研究与开发;孙辉;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅰ辑》;20190315(第03期);第B016-68页，正文第26、49-50、65-68页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113989808A (en) | 2022-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100414463C (en) | Integrated support system for supporting sheet metal machining | |
| CN101419612B (en) | Image processing device and image processing method | |
| CN113989808B (en) | Method and system for selecting specifications of mechanical manufacturing materials based on drawing information processing | |
| CN113435773B (en) | Production progress monitoring method, system and storage medium for digital factory | |
| US20170323150A1 (en) | Object formation image management system, object formation image management apparatus, and non-transitory computer readable medium | |
| CN114186298A (en) | Intelligent mechanical part manufacturing and operating method and system | |
| CN116245882A (en) | Circuit board electronic element detection method and device and computer equipment | |
| CN112434488A (en) | Automatic marking method, equipment and computer readable storage medium | |
| US10549349B2 (en) | Method and apparatus for controlling a 3D-printing device and 3D-printing device | |
| CN111881171B (en) | Drawing identification recommendation method and system based on data analysis | |
| CN111401438B (en) | Image sorting method, device and system | |
| CN111126030B (en) | Label typesetting processing method, device and system | |
| CN110443306B (en) | Authenticity identification method for wine cork | |
| CN111160374A (en) | Color identification method, system and device based on machine learning | |
| CN109063708B (en) | Industrial image feature identification method and system based on contour extraction | |
| CN107315392B (en) | Special-shaped workpiece discharging optimization algorithm of automatic feeder of punch press and tolerance algorithm thereof | |
| CN113781432B (en) | Laser scanning automatic laying on-line detection method and device based on deep learning | |
| CN114926842A (en) | Dongba pictograph recognition method and device | |
| JP2016091501A (en) | Blank product discrimination system, blank product discrimination apparatus, blank product discrimination method, and program | |
| CN115526859A (en) | Method for identifying production defects, distributed processing platform, equipment and storage medium | |
| CN115958792A (en) | Printing quality analysis method and system of 3D printer | |
| JP4755779B2 (en) | 3D inspection and utilization method for sheet metal products and its system | |
| CN111169164B (en) | Apparatus and method for inspecting printing of articles | |
| CN112784620A (en) | Tracing method and tracing system for product manufacturing | |
| CN116977998B (en) | Workpiece feeding visual identification system and method for coating production line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 523078 Building 3, No. 2, Xinke Road, Nancheng Street, Dongguan City, Guangdong Province Patentee after: Guangdong Saide Automation Technology Co.,Ltd. Address before: 523087 d2E, Xinke factory area, Hongyuan Industrial Zone, Xinke Road, Nancheng street, Dongguan City, Guangdong Province Patentee before: GUANGDONG HONGYUAN XINKE AUTOMATION TECHNOLOGY DEVELOPMENT Co.,Ltd. |