WO2020007177A1 - 计算机执行的报价方法、报价装置、电子设备及存储介质 - Google Patents

计算机执行的报价方法、报价装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2020007177A1
WO2020007177A1 PCT/CN2019/091122 CN2019091122W WO2020007177A1 WO 2020007177 A1 WO2020007177 A1 WO 2020007177A1 CN 2019091122 W CN2019091122 W CN 2019091122W WO 2020007177 A1 WO2020007177 A1 WO 2020007177A1
Authority
WO
WIPO (PCT)
Prior art keywords
product
appearance
structural
parameters
feature
Prior art date
Application number
PCT/CN2019/091122
Other languages
English (en)
French (fr)
Inventor
胡拯纲
侯健
Original Assignee
京东方科技集团股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 京东方科技集团股份有限公司 filed Critical 京东方科技集团股份有限公司
Priority to JP2020568293A priority Critical patent/JP7405775B2/ja
Priority to US16/633,214 priority patent/US11544751B2/en
Priority to EP19830747.2A priority patent/EP3819854A4/en
Publication of WO2020007177A1 publication Critical patent/WO2020007177A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • G06Q10/0875—Itemisation or classification of parts, supplies or services, e.g. bill of materials
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0611—Request for offers or quotes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/211—Selection of the most significant subset of features
    • G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0283—Price estimation or determination
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0623—Item investigation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/771—Feature selection, e.g. selecting representative features from a multi-dimensional feature space
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/19—Recognition using electronic means
    • G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06V30/19147—Obtaining sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/19—Recognition using electronic means
    • G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06V30/19173—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/40—Document-oriented image-based pattern recognition
    • G06V30/42—Document-oriented image-based pattern recognition based on the type of document
    • G06V30/422—Technical drawings; Geographical maps

Definitions

  • the present disclosure relates to the technical field related to product quotation, and particularly to a computer-implemented quotation method, quotation device, electronic device, and storage medium.
  • the quotation process of more general B2B products is as follows: first, the customer requests the appearance and electrical specifications of the product, and then the business staff of the company analyzes the customer's needs and confirms the feasibility with the R & D personnel, and then according to the appearance of the product Specifications, use software tools to draw pictures of product appearance. Finally, the business staff collaborates with the R & D staff and selects a similar historical product based on experience, calculates its BOM (Material of Material) cost, plus other factors that affect the cost, and comprehensively obtains the product quotation.
  • BOM Standard of Material
  • a computer-implemented quotation method including: obtaining structural parameters and electrical parameters of a product; using the structural parameters of the product to construct an appearance diagram of the product, and comparing the appearance diagram and history of the product
  • Product appearance diagrams are compared for similarity ranking to obtain appearance similarity ranking
  • product electrical parameters are compared with historical product electrical parameters for similarity ranking to obtain electrical parameter similarity ranking
  • Similarity ranking with electrical parameters results in comprehensive ranking based on structural parameters and electrical parameters
  • comparing the appearance of the product with the appearance of historical products to obtain similarity rankings includes: detecting the appearance of the product to extract structural features in the appearance of the product; using classification The device sorts the similarity between the structural features in the appearance diagram of the historical product and the structural features in the appearance diagram of the product to obtain the appearance similarity ranking.
  • the method further includes: performing signal transformation and noise reduction pre-processing on the appearance of the product.
  • detecting the appearance of the product includes: using a scanning sub-window to move in the appearance of the product; for each position in the appearance of the product determined during the movement of the scanning sub-window, calculating the Location characteristics.
  • using a classifier to sort similarities between structural features in the appearance map of historical products and structural features in the appearance map of products includes: using a classifier to recursively eliminate algorithms based on features in the appearance maps of historical products.
  • the structural features are sorted by similarity with the structural features in the product's appearance diagram.
  • Permutation score remove the feature with the lowest permutation score in the initial feature arrangement, and update to obtain a new feature arrangement; repeat the above-mentioned cycle until the feature arrangement includes only one feature, and obtain the similarity ranking of the structural features based on the order of feature removal. .
  • the method further includes training the classifier using training sample data, based on a knowledge base, or constraints, wherein the training sample data includes positive samples and negative samples, and the positive samples are to be detected.
  • a sample of a structural feature, the negative sample is a sample that does not include a structural feature to be detected.
  • the appearance drawing includes the shape, size, material, and design parameters of the product, and the appearance drawing is a six-view drawing of the product drawn at the same scale.
  • comparing the electrical parameters of the product with the electrical parameters of the historical product to obtain a similarity ranking of the electrical parameters includes: determining that the historical product has the same number of electrical parameters as the product; based on the number of the same electrical parameter, Sort historical products.
  • a comprehensive ranking based on structural parameters and electrical parameters is calculated based on the cost weights of structural components and electrical components
  • the appearance similarity ranking and electrical parameter similarity ranking calculations include: determining the appearance based on the cost weights of structural components and electrical components Weights of similarity ranking and electrical parameter similarity; weight calculation based on the appearance similarity ranking and electrical parameter similarity calculation to obtain a comprehensive ranking based on structural parameters and electrical parameters.
  • determining the BOM of the product further includes: adjusting the BOM of the product based on the structural parameters and electrical parameters of the product to obtain a BOM of the product that conforms to the product's structural parameters and electrical parameters.
  • the product quote is calculated by the following formula:
  • Product quotation ⁇ (component cost) ⁇ processing rate ⁇ other rates; of which, the cost of parts and components is calculated based on the bill of materials of the product, and the processing rate and other rates are obtained from the enterprise resource planning system.
  • a product quotation device including: a parameter obtaining unit configured to obtain structural parameters and electrical parameters of the product; an appearance similarity ranking unit configured to construct using the product's structural parameters The appearance diagram of the product, comparing the appearance diagram of the product with the appearance diagram of the historical product to obtain the appearance similarity ranking; the electrical parameter similarity ranking unit is configured to compare the electrical parameters of the product with the electrical parameters of the historical product The similarity comparison is performed to obtain the electrical parameter similarity ranking; the comprehensive ranking unit is configured to obtain a comprehensive ranking based on structural parameters and electrical parameters based on the appearance weight similarity ranking and electrical parameter similarity ranking based on the cost weights of structural components and electrical components; and products; and The quoting unit is configured to determine a bill of materials of the products based on the comprehensive sorting, and calculate a quote of the products based on the bill of materials of the products.
  • an electronic device including: at least one processor; and a memory communicatively connected with the at least one processor; wherein the memory stores a process that can be processed by the at least one
  • the instructions executed by the processor enable the at least one processor to execute the computer-implemented quotation method as described above.
  • a computer-readable storage medium having stored thereon instructions that, when executed by a processor, cause the processor to execute the computer-implemented quotation method as described above .
  • FIG. 1A is a schematic flowchart of a computer-implemented quotation method according to an embodiment of the present disclosure
  • FIG. 1B is a schematic flowchart of an embodiment of a computer-implemented quotation method provided by the present disclosure
  • FIG. 2 is a schematic block diagram of a computer-implemented quotation method provided by the present disclosure
  • FIG. 3 is a schematic diagram of feature extraction and classifier training provided by the present disclosure
  • FIG. 4 is a schematic diagram of feature ranking calculation provided by the present disclosure.
  • FIG. 5 is a schematic structural block diagram of a quotation device provided by the present disclosure.
  • FIG. 6 is a schematic diagram of a hardware structure of a device for performing a quotation method provided by the present disclosure
  • FIG. 7 illustrates a schematic diagram of an architecture of an exemplary computing device according to an embodiment of the present disclosure
  • FIG. 8 shows a schematic diagram of a storage medium according to an embodiment of the present disclosure.
  • the purpose of the present invention is to propose a computer-implemented quotation method, quotation device and electronic equipment, which can quickly and accurately implement product quotation, and greatly reduce the dependence of product quotation on R & D engineers, and further reduce the labor and material costs of product quotation.
  • FIG. 1A shows a schematic flowchart of a computer-implemented quotation method according to an embodiment of the present disclosure.
  • the quotation method can be used to implement product quotation.
  • the product may be, for example, a new product that needs to provide a quotation.
  • a computer may calculate a product quotation by combining the structural parameters and electrical parameters of the product (ie, the new product) and the parameters of the historical product. While increasing the quotation rate and accuracy, it avoids the labor and material consumption of manual quotation.
  • the structural parameters and electrical parameters of the product are acquired.
  • the structural parameters and electrical parameters corresponding to the product can be obtained based on the customer's demand for the product.
  • the structural parameters refer to parameters related to the structural design of the product.
  • the structural parameters can be divided into a plurality of main materials, such as a front frame, a screen, a rear case, a button, a bracket, and a base.
  • attribute parameters related to each major material such as shape, size, material, appearance, etc.
  • the electrical parameters are parameters related to the electrical design of the product.
  • the electrical parameters may include the key technology, hardness, brightness, board power, size, etc.
  • the obtaining process may correspond to a user inputting the structural parameters and electrical parameters to a processor such as a computer through an input device.
  • the computer may be stored in a local or cloud storage, for example.
  • the computer may also obtain the parameter based on a wireless network through a wireless device.
  • the structural appearance parameters of the products are constructed using the structural parameters of the products, and the similarities of the appearance appearances of the products and the historical appearance appearance products are compared to obtain the appearance similarity ranking.
  • the computer may generate an appearance drawing of the product by drawing software installed therein based on the structural parameters of the product.
  • Appearance drawings of the historical product may be stored, for example, in a database accessible by the computer.
  • the appearance of the product is used to reflect the overall performance of the product's size, external structure, color, pattern, shape, etc., and can be used to reflect the quality of the product.
  • the appearance chart constructed using the structural parameters of the product can be six views of the product drawn at the same scale, so that the appearance chart can intuitively reflect the structural parameters such as the shape, size, material, and design of the product.
  • the appearance diagram constructed using the structural parameters of the product may also be a three-dimensional three-dimensional structure diagram, which is used to stereoscopically display the structural parameters such as the shape, size, material, and design of the product.
  • the appearance similarity ranking is used to reflect the appearance similarity between historical products and products, and the appearance may be reflected by structural features.
  • signal transformation and noise reduction pre-processing may be performed on the appearance diagram of the product to remove impurities and interference factors in the signal, thereby improving the accuracy of subsequent recognition.
  • the signal conversion may include converting an image signal of an appearance image into an electrical signal.
  • comparing the appearance of the product with the appearance of the historical product to obtain the appearance similarity ranking may include: detecting the appearance of the product to extract structural features in the appearance of the product, wherein Detecting the appearance of the product includes: using a scanning sub-window to move the appearance of the product; for each position in the appearance of the product determined during the movement of the scanning sub-window, calculating the structural characteristics of the position.
  • comparing the appearance map of the product with the appearance map of the historical product to obtain the appearance similarity ranking may further include: using a classifier to compare the structural features in the appearance map of the historical product with the appearance map of the product. Structural features are sorted by similarity to obtain appearance similarity.
  • a classifier may use a feature recursive elimination algorithm to sort similarities between the structural features in the appearance map of the historical product and the structural features in the appearance map of the product.
  • Permutation score remove the feature with the lowest permutation score in the initial feature arrangement, and update to obtain a new feature arrangement; repeat the above-mentioned cycle until the feature arrangement includes only one feature, and obtain the similarity ranking of the structural features based on the order of feature removal .
  • the computer-implemented quoting method may further train the classifier, and the training may include training the classifier using training sample data, based on a knowledge base, or constraints, where the training The sample data includes positive samples and negative samples, the positive samples are samples containing structural features to be detected, and the negative samples are samples not containing structural features to be detected.
  • step S103 the electrical parameters of the product are compared with the historical products' electrical parameters to obtain a similarity ranking of the electrical parameters.
  • the appearance similarity ranking based on the comparison of the product's appearance map with the historical product's appearance map involves the structural characteristics of the mechanical parts of the product.
  • the key factors affecting the product cost include the liquid crystal panel and internal circuit devices. Wait. Based on these characteristics, a list of electrical parameters corresponding to the product can be established for comparing the similarity between the electrical parameters of the product and the electrical parameters of the historical product to obtain a similarity ranking of the electrical parameters.
  • the electrical parameter similarity ranking is used for Reflects the similarity between the historical product and the product in electrical parameters. The higher the ranking, the closer the historical product is to the product in terms of electrical parameters.
  • the product may include, for example, 10 electrical parameters.
  • the historical product 1 has 8 electrical parameters that are the same as the electrical parameters of the product
  • the historical product 2 has 6 electrical parameters that are the same as the product
  • the historical product 3 has 5 electrical parameters that are the same as the electrical parameters of the product, so that the historical products can be sorted based on the same number of electrical parameters.
  • the resulting ranking can be, for example, historical product 1, historical product 2, Historical products 3.
  • step S104 a comprehensive ranking based on the structural parameters and electrical parameters is obtained based on the cost weights of the structural components and electrical components, the appearance similarity ranking, and the electrical parameter similarity ranking.
  • a comprehensive ranking based on the structural parameters and electrical parameters is obtained based on the cost weights of the structural components and electrical components, the appearance similarity ranking, and the electrical parameter similarity ranking.
  • the weights of the appearance similarity order and the electrical parameter similarity may be determined based on the cost weights of the structural parts and the electrical components, and the weights based on the appearance similarity order and the electrical parameter similarity are calculated to obtain the structural parameter and electrical Comprehensive ordering of parameters.
  • the cost weights can be averaged to determine the appearance similarity order and electrical parameters through the appearance similarity order and the electrical parameter similarity order. Similarity weights are then obtained based on a comprehensive ranking of structural parameters and electrical parameters.
  • a bill of materials of the products is determined based on the comprehensive sorting, and the product quotation is calculated based on the bill of materials of the products.
  • the bill of materials (BOM) of the product refers to the parts list and structure required for the product.
  • the BOM of the historical product with the highest ranking can be used as the BOM of the product based on the comprehensive sorting.
  • its bill of materials can be stored in a computer-accessible database along with the historical product, such as electrical parameters and structural parameters. After the historical product with the most comprehensive ranking is determined, the computer can directly access the database and propose a physical list of the historical products.
  • the bill of materials is a list of all sub-assemblies, parts, and raw materials constituting the assembly, as well as a list of the quantities of each component required to make an assembly.
  • the BOM can be associated with the unit price of each part of the product.
  • a quotation for the product is then calculated based on the product's bill of materials. For example, the quotation of the product is determined based on the number of various parts included in the product's bill of materials and the unit price.
  • determining the BOM of the product may further include: adjusting the BOM of the product based on the structural parameters and electrical parameters of the product to obtain a BOM of the product that conforms to the product's structural parameters and electrical parameters. . That is, if the obtained product bill of materials still cannot fully meet all the product parameter requirements, a certain degree of modification or addition or deletion of the bill of materials is needed to make the components in the bill of materials fully meet the product's structural parameters and Requirements for electrical parameters. The quotation for the product can then be determined based on the adjusted BOM of the product.
  • the product quote is calculated by the following formula:
  • the cost of parts and components is calculated based on the bill of materials of the product, and the processing rate and other rates are obtained from the enterprise resource planning system.
  • the processing rate may include equipment loss, processing cycle, and the like, such as machine tonnage, machine cost / hour, and product molding cycle.
  • the other rates may include management costs, transportation costs, packaging costs, and so on.
  • processing rates and other rates are generally available in the enterprise resource planning system (ERP).
  • ERP enterprise resource planning system
  • FIG. 1B is a schematic flowchart of an embodiment of a quotation method provided by the present disclosure. The quotation method according to an embodiment of the present disclosure will be described in detail below with reference to FIG. 1B.
  • step S1 the structural parameters and electrical parameters of the product are obtained based on the customer's demand for the product.
  • step S2 the structural appearance parameters of the products are constructed by using the structural parameters of the products, and the similarity comparison between the external appearance drawings of the products and the historical appearance products is performed to obtain the appearance similarity ranking.
  • the step S2 may include steps S21-S23.
  • step S21 the appearance of the product is detected and the structural features in the appearance of the product are extracted accordingly based on the product type and structural features of the product.
  • scanning the sub-window can be used to achieve the detection of the appearance, which specifically includes: first, using the scanning sub-window to continuously shift and slide in the appearance of the product to be tested; second, during the shift and sliding process of the product
  • the structural characteristics of the location area are calculated by scanning the sub-windows for each different position in the appearance map of.
  • the appearance of the product may be subjected to signal conversion and noise reduction pre-processing.
  • step S22 the extracted structural features are input to a pre-built classifier for classification processing, and the ranking of each structural feature is obtained.
  • the ordering of each structural feature refers to, for each of the respective structural features, determining the ordering of historical products and the structural feature of the product. In other words, an ordering based on each structural feature can be obtained in step S22.
  • a feature recursive elimination algorithm can be used to calculate the ranking of structural features.
  • the specific steps include:
  • w is the weight corresponding to the structural feature
  • m is the total number of historical products with the structural feature
  • ⁇ i is the weight ratio
  • (x i , y i ) is the coordinate value used to represent the structural feature of the image. Represents image characteristics.
  • the method for converting the parameter values corresponding to the structural features into coordinate values can be any currently feasible transformation scheme, and this embodiment is not specifically limited.
  • the permutation score corresponding to each structural feature is calculated based on the weight corresponding to each structural feature; it is calculated by the following formula:
  • c j (w j ) 2 ; wherein w j is the weight of the j-th structural feature; c j is the arrangement score corresponding to the j-th structural feature.
  • the structure feature order is obtained accordingly.
  • FIG. 4 is a schematic diagram of a calculation principle of structural feature ranking provided by the present disclosure.
  • the feature selection algorithm used in the above embodiment of the present application is a feature recursive elimination algorithm (SVM-RFE) based on a Support Vector Machine.
  • SVM-RFE feature recursive elimination algorithm
  • SVM-RFE is based on 2-norm SVM, and its mathematical model is:
  • the loop process is executed until there is only one feature left in the feature set, and as a result, a list of structural feature sequence numbers sorted by feature importance is obtained.
  • the SVM-RFE algorithm is a backward search algorithm.
  • the redundant structural features are first removed, and then a new set of ranking scores are iteratively calculated again. Until a subset of the structural features that have the greatest influence on the classification result is left, the purpose of reducing the dimension of the structural features and improving the classification accuracy can be achieved.
  • step S23 for different historical products, based on the ranking of the product structural features and the corresponding structural features in the historical products, the similarity ranking of the historical products based on the structural features is comprehensively calculated to obtain the similar appearance of the historical products based on the products. Degree sort.
  • the step of calculating the similarity ranking of historical products based on structural features may further include:
  • the structural features of the current historical product and the corresponding order are extracted from the order of the product structural features.
  • the similarity between the current historical product and the product is obtained through the calculation of the structural characteristics of the historical product and the corresponding calculation.
  • a comprehensive ranking of historical products relative to product structural characteristics can be obtained, that is, the appearance similarity ranking of all historical products and current products can be obtained.
  • a classifier may use a feature recursive elimination algorithm to sort similarities between the structural features in the appearance map of the historical product and the structural features in the appearance map of the product.
  • FIG. 3 is a schematic diagram of feature extraction and classifier training provided by the present disclosure. It can be known from FIG. 3 that before the above-mentioned classification is implemented by the classifier, the classifier needs to be trained.
  • the training sample data of the classifier includes a positive sample and a negative sample, wherein the positive sample is a sample including a target to be inspected, and the negative sample is a sample not including a target to be inspected.
  • the classifier can be trained based on the knowledge base or constraints. Because the sample data has a large amount of data, and thus the extracted feature has a very large amount of data, in order to shorten the training process for the classifier, a knowledge base (for example, a rule) can be added or restrictions can be introduced to narrow the search range. For example, through the knowledge base, it is possible to determine that a certain position of a TV picture represents a high probability of pressing a key, so that the search range can be greatly reduced.
  • a knowledge base for example, a rule
  • step S3 the electrical parameters of the product are compared with the electrical parameters of the historical product to obtain a similarity ranking of the electrical parameters.
  • step S4 based on the difference in cost weights of the structural components and electrical components in the historical product, a comprehensive ordering of the historical parameters corresponding to the historical parameters is obtained by calculating the appearance similarity order and the electrical parameter similarity order.
  • step S5 based on comprehensively sorting the previous historical product's bill of materials, the bill of materials for the products is extracted and constructed, and the product quotation of the products is calculated accordingly.
  • the step of extracting the BOM of the product further includes: adjusting and correcting the BOM of the product based on the structural parameters and electrical parameters of the product, so as to make the BOM of the product Meet the product's structural and electrical parameters. That is, if the obtained product bill of materials still cannot fully meet all the product parameter requirements, a certain degree of modification or addition or deletion of the bill of materials is needed to make the components in the bill of materials fully meet the product's structural parameters and Requirements for electrical parameters.
  • the cost of parts and components is mainly the price of raw materials, including: specifications of raw materials, prices of raw materials, net weight of products, glue ports, and losses.
  • the product appearance map is obtained by constructing the product's structural parameters accordingly, and then the appearance similarity ranking can be obtained by identifying and comparing with the historical product appearance map (such as through a classifier), while using Compare the electrical parameters to get the similarity ranking of the historical products, and then use the comprehensive analysis of the two rankings to get the comprehensive ranking based on the structural parameters and electrical parameters of the historical products, so that the historical products that are similar to the products can be directly sorted.
  • the extraction will further facilitate the subsequent calculation of product quotations by using the historical product bill of materials.
  • the above-mentioned sorting based on the comparison of specific parameters will be more accurate and computational efficiency will be better than manual screening. Accuracy and quote rate. Therefore, the method disclosed in the present disclosure can not only quickly and accurately implement product quotation, but also greatly reduce the dependence of product quotation on R & D engineers, and further reduce the labor and material costs of product quotation.
  • FIG. 2 is a schematic block diagram of a computer-implemented quotation method provided by the present disclosure.
  • the quotation method according to the present disclosure may include two main parts. One is to compare the appearance of the product with the appearance of the historical product through image recognition technology to obtain the appearance similarity ranking. For example, image recognition Technical comparison of the appearance of historical products with the appearance of products. The second is to compare the electrical parameters of historical products and the electrical parameters of products to get the similarity ranking of electrical parameters. Based on the appearance similarity ranking and the electrical parameter similarity ranking, the comprehensive similarity ranking of historical products is obtained. Next, select the historical products that will be ordered in the comprehensive similarity order, and call up their BOM and corresponding price information as the product's bill of materials. Finally, the cost of each component in the product BOM is calculated according to the formula to obtain an estimated quotation.
  • the problems that the quotation method provided in this disclosure can solve include: (1) the current problem of unscientific and inaccurate product quotations can be solved; (2) breaking the barriers of sales and R & D departments, reducing the communication time, and quickly querying similar products as Quotation reference; (3) accurate product quotations can be obtained through scientific calculation methods; (4) help enterprises to achieve successful bidding as soon as possible while obtaining maximum profits; (5) reduce the workload of artificially querying historical product price information.
  • the quotation method, quotation device, and electronic equipment provided by the present disclosure can obtain the product's appearance map by constructing the product's structural parameters, and then can obtain the appearance similarity ranking by performing similarity comparison with the historical product's appearance map, and use electrical parameter comparison to obtain The electrical parameter similarity is sorted, and then the comprehensive sorting of historical products is obtained through the comprehensive analysis of the two sorts, so as to extract historical products that are similar to the products, and use the bill of materials of historical products to calculate the product quotation accordingly.
  • the above-mentioned sorting obtained by comparison based on specific parameters will be more accurate and computationally more efficient, and will have a higher accuracy rate and a faster quote rate than manual comparison. Therefore, this application can not only quickly and accurately implement product quotations, but also greatly reduce the dependence of product quotations on R & D engineers, and further reduce the labor and material costs of product quotations.
  • the product quotation device includes a parameter obtaining unit 1, an appearance similarity ranking unit 2, an electrical parameter similarity ranking unit 3, a comprehensive ranking unit 4, and a product pricing unit 5.
  • the parameter acquisition unit 1 may be configured to acquire structural parameters and electrical parameters of the product. And the parameter obtaining unit 1 may be further configured to send the obtained structural parameters and electrical parameters to the appearance similarity ranking unit 2 and the electrical parameter similarity ranking unit 3.
  • the appearance similarity ranking unit 2 may be configured to use the structural parameters of the product to construct the appearance map of the product, and compare the appearance map of the product with the appearance map of the historical product to obtain the appearance similarity ranking.
  • the electrical parameter similarity ranking unit 3 may be configured to compare the electrical parameters of the product with the electrical parameters of the historical product to obtain a similarity ranking of the electrical parameters.
  • the comprehensive sorting unit 4 may be configured to obtain the comprehensive sorting based on the structural parameters and electrical parameters based on the cost weights of the structural components and electrical components, the appearance similarity ranking, and the electrical parameter similarity ranking.
  • the product quoting unit 5 may be configured to determine a bill of materials of the products based on the comprehensive sorting, and calculate a quote of the products based on the bill of materials of the products.
  • the product quotation device can compare existing historical products based on the product appearance map and electrical parameters, and obtain a bill of materials for historical products with a high comprehensive similarity ranking as a bill of materials for the product to calculate a new product quote.
  • the calculated product quotation can be sent to the market side, to quickly estimate the effect of the quotation, reduce the difficulty and workload of communication between the sales side and the R & D side, and further, allow customers to quickly obtain products Quotation, thus speeding up order negotiation.
  • FIG. 6 is a schematic diagram of a hardware structure of an electronic device performing a quotation method provided by the present disclosure.
  • the electronic device includes: at least one processor 201 and a memory 202.
  • One processor 201 is taken as an example in FIG. 6.
  • the memory 202 stores instructions executable by the at least one processor, so that the at least one processor can execute the quotation method as described above.
  • the electronic device that executes the quotation method may further include an input device 203 and an output device 204.
  • the processor 201, the memory 202, the input device 203, and the output device 204 may be connected through a bus or other methods. In FIG. 6, the connection through the bus is taken as an example.
  • the memory 202 is a non-volatile computer-readable storage medium and can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as program instructions corresponding to the quotation method in the embodiment of the present application. Module.
  • the processor 201 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions, and modules stored in the memory 202, that is, the quotation method of the foregoing method embodiment is implemented.
  • the memory 202 may include a storage program area and a storage data area, where the storage program area may store an operating system and application programs required for at least one function; the storage data area may store data created according to the use of the product quotation device, and the like.
  • the memory may include read-only memory (ROM), random-access memory (RAM), or other optical disk storage, magnetic disk storage, or other magnetic storage devices, or can be used to carry or store instructions or data structures Any storage medium of the form that expects program code and that can be accessed by a computer.
  • the memory 202 may include a high-speed random access memory, and may further include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage device.
  • the memory 202 may optionally include a memory remotely disposed with respect to the processor 201, and these remote memories may be connected to the product quotation device through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
  • the input device 203 can receive inputted numeric or character information, and generate key signal inputs related to user settings and function control of the product quotation device.
  • the output device 204 may include a display device such as a display screen.
  • the one or more modules are stored in the memory 202, and when executed by the one or more processors 201, perform the quotation method in the foregoing embodiment.
  • the electronic device in the embodiment of the present disclosure may exist in various forms, including but not limited to:
  • Mobile communication equipment This type of equipment is characterized by mobile communication functions, and its main goal is to provide voice and data communication.
  • Such terminals include: smart phones (such as iPhone), multimedia phones, feature phones, and low-end phones.
  • Ultra-mobile personal computer equipment This type of equipment belongs to the category of personal computers, has computing and processing functions, and generally has mobile Internet access characteristics.
  • Such terminals include: PDA, MID and UMPC devices, such as iPad.
  • Portable entertainment equipment This type of equipment can display and play multimedia content. These devices include audio and video players (such as iPods), handheld game consoles, e-books, as well as smart toys and portable car navigation devices.
  • Server A device that provides computing services.
  • the composition of the server includes processors, hard disks, memory, and system buses.
  • the server is similar to a general-purpose computer architecture. , Reliability, security, scalability, manageability and other aspects of higher requirements.
  • the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, and an input / output component 3060. , Hard disk 3070 and so on.
  • a storage device in the computing device 3000, such as the ROM 3030 or the hard disk 3070, may store various data or files used for processing and / or communication of the quotation method provided by the present disclosure and program instructions executed by the CPU.
  • the computing device 3000 may also include a user interface 3080.
  • FIG. 7 is only exemplary. When implementing different devices, one or more components in the computing device shown in FIG. 7 may be omitted according to actual needs. According to an embodiment of the present disclosure, there is also provided a computer-readable storage medium having stored thereon instructions that, when executed by a processor, cause the processor to execute the quotation method as described above.
  • FIG. 8 shows a schematic diagram 4000 of a storage medium according to the present disclosure.
  • the computer storage medium 4020 stores computer-readable instructions 4010.
  • the quotation method according to an embodiment of the present disclosure described with reference to the above drawings may be executed.
  • the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory, for example.
  • the volatile memory may include, for example, a random access memory (RAM) and / or a cache memory.
  • the non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, and the like.
  • DRAM dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Development Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Evolutionary Computation (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Image Analysis (AREA)

Abstract

公开了一种计算机执行的报价方法，包括：获取所述产品的结构参数和电气参数(S101)；利用产品的结构参数构建所述产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序(S102)；将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序(S103)；基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序(S104)；以及基于所述综合排序确定所述产品的物料清单，基于所述产品的物料清单计算所述产品报价(S105)。

Description

计算机执行的报价方法、报价装置、电子设备及存储介质 技术领域
本公开涉及产品报价相关技术领域，特别是指一种计算机执行的报价方法、报价装置、电子设备及存储介质。
背景技术
当前，较为通用的B2B产品(例如显示产品)报价过程为：首先由客户提出对产品的外观和电气规格要求，然后企业业务人员通过分析客户需求、与研发人员确认可行性后，根据产品的外观规格，利用软件工具绘制产品外观图片。最后，业务人员与研发人员协作并根据经验选择一款类似的历史产品，计算其BOM(Bill of Material，物料清单)成本，加上其他影响成本的因素，综合得出产品的报价。但是当前使用的产品报价方法不仅对研发工程师的依赖性过大，而且这样的选取方式难以遍历全部历史产品，导致报价周期过长，同时也必定使得产品的报价不够精准。
发明内容
根据本公开的一方面，提供了一种计算机执行的报价方法，包括：获取产品的结构参数和电气参数；利用产品的结构参数构建所述产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序；将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序；基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序；以及基于所述综合排序确定所述产品的物料清单，并基于所述产品的物料清单计算所述产品的报价。
根据本公开实施例，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序包括：对产品的外观图进行检测以提取产品的外观图中的结构特征；利用分类器对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序，以得到外观相似度排序。
根据本公开实施例，所述方法还包括：对产品的外观图进行信号变换和降噪预处理。
根据本公开实施例，对产品的外观图进行检测包括：利用扫描子窗口在产品的外观图中进行移动；对于扫描子窗口在移动过程中确定的产品的外观图中的每个位置，计算该位置的特征。
根据本公开实施例，利用分类器对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序包括：利用分类器基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序。
根据本公开实施例，基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序包括：将结构特征对应参数值转化为坐标值，得到初始特征排列；计算每个结构特征对应的权重，通过如下公式计算所述权重：
Figure PCTCN2019091122-appb-000001
其中，w为结构特征对应的权重，m为具有该结构特征的历史产品的总数，α i为权重比，(xi，yi)为用于表示图像结构特征的坐标值；基于每个结构特征对应的权重计算每个结构特征对应的排列分数，通过如下公式计算：c j＝(w j) 2；其中，w j为第j个结构特征对应的权重，c j为第j个结构特征对应的排列分数；在初始特征排列中去除排列分数最小的特征，并更新得到新的特征排列；重复上述循环过程直到所述特征排列中只包括一个特征，基于特征去除的顺序得到结构特征的相似度排序。
根据本公开实施例，所述方法还包括利用训练样本数据、基于知识库或者限制条件训练所述分类器，其中，所述训练样本数据包括正样本和负样本，所述正样本为包含待检测结构特征的样本，所述负样本为不包含待检测结构特征的样本。
根据本公开实施例，所述外观图中包括产品的形状、尺寸、材质以及外观设计参数，且所述外观图为采用同一比例绘制的产品六视图。
根据本公开实施例，将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序包括：确定历史产品具有与产品相同的电气 参数的数目；基于相同的电气参数的数目，对历史产品进行排序。
根据本公开实施例，基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序计算得到基于结构参数和电气参数的综合排序包括：基于结构部件和电气元件的成本权重确定外观相似度排序和电气参数相似度的权重；基于所述外观相似度排序和电气参数相似度的权重计算得到基于结构参数和电气参数的综合排序。
根据本公开实施例，确定所述产品的物料清单还包括：基于所述产品的结构参数和电气参数，对产品的物料清单进行调整，得到符合产品的结构参数和电气参数的产品的物料清单。
根据本公开实施例，所述产品报价通过如下公式计算：
产品报价＝Σ(零部件成本)×加工费率×其他费率；其中，零部件成本是基于所述产品的物料清单计算得到，加工费率和其他费率是从企业资源计划系统中得到。
根据本公开的另一方面，还提供了一种产品报价装置，包括：参数获取单元，配置成获取所述产品的结构参数和电气参数；外观相似度排序单元，配置成利用产品的结构参数构建所述产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序；电气参数相似度排序单元，配置成将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序；综合排序单元，配置成基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序；以及产品报价单元，配置成基于所述综合排序确定所述产品的物料清单，并基于所述产品的物料清单计算所述产品的报价。
根据本公开的又一方面，还提供了一种电子设备，包括：至少一个处理器；以及与所述至少一个处理器通信连接的存储器；其中，所述存储器存储有可被所述至少一个处理器执行的指令，使所述至少一个处理器能够执行如上所述的计算机执行的报价方法。
根据本公开的又一方面，还提供了一种计算机可读存储介质，其上存储有指令，所述指令在被处理器执行时，使得所述处理器执行如上所述的计算机执行的报价方法。
附图说明
图1A为根据本公开实施例的计算机执行的报价方法的示意流程图；
图1B为本公开提供的计算机执行的报价方法一个实施例的流程示意图；
图2为本公开提供的计算机执行的报价方法的示意性框图；
图3为本公开提供的特征提取以及分类器训练的示意图；
图4为本公开提供的特征排序计算示意图；
图5为本公开提供的报价装置的示意性结构框图；
图6为本公开提供的执行报价方法的设备的硬件结构示意图；
图7示出了根据本公开实施例的示例性计算设备的架构的示意图；
图8示出了根据本公开实施例的存储介质的示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚明白，以下结合具体实施例，并参照附图，对本发明进一步详细说明。
需要说明的是，本公开实施例中所有使用“第一”和“第二”的表述均是为了区分两个相同名称非相同的实体或者非相同的参量，可见“第一”“第二”仅为了表述的方便，不应理解为对本公开实施例的限定，后续实施例对此不再一一说明。
本发明的目的在于提出一种计算机执行的报价方法、报价装置及电子设备，能够快速准确的实现产品报价，而且大大降低了产品报价对于研发工程师的依赖，进一步降低了产品报价的人力物力成本。
图1A示出了根据本公开实施例的计算机执行的报价方法的示意性流程图。所述报价方法可以用于实现产品报价。所述产品例如可以是一种需要提供报价的新产品，根据本公开的报价方法，计算机可以结合所述产品(即，新产品)的结构参数和电气参数以及历史产品的参数来计算产品报价，提高了报价速率、准确性的同时，避免了人工报价的人力物力消耗。
首先，在步骤S101，获取产品的结构参数和电气参数。例如，可以基于客户对产品的需求来获取与产品相应的结构参数和电气参数。其中，所述结构参数是指对于产品的结构设计相关的参数，例如以电视机产品为例，结构参数可以分为多个主要材料，诸如前框、屏幕、后壳、按键、支架、底座， 以及与各个主要材料相关的属性参数，诸如形状、尺寸、材质、外观等。所述电气参数是与产品的电气设计相关的参数，例如，以产品为显示产品为例，电气参数可以包括显示产品的液晶面板的关键技术、硬度、亮度、板卡功率、尺寸等。此外，对于一些产品，可能存在一些参数既不属于结构参数也不属于电气参数，或者同时具有结构参数和电气参数特点，此时可以预先将其归入到结构参数或电气参数中的一类以实现统一的划分。举例来说，所述获取过程可以对应于用户通过输入设备来向诸如计算机等的处理器输入所述结构参数和电气参数。所述计算机在接收到所述参数后，例如可以存储在本地或者云端存储器中。或者，所述计算机还可以通过无线设备基于无线网络来获取所述参数。
接着，在步骤S102，利用产品的结构参数构建产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序。例如，所述计算机可以基于产品的结构参数来通过安装在其内的绘图软件来生成所述产品的外观图。所述历史产品的外观图例如可以存储在所述计算机可访问的数据库中。其中，产品的外观图用于反映产品的大小、外部结构、颜色、图案、形状等方面的综合表现，可以用于反映产品的质量。根据本公开实施例，利用产品的结构参数构建的外观图既可以是采用同一比例绘制的产品六视图，使得外观图可以直观地体现产品的形状、尺寸、材质以及外观设计等结构参数。根据本公开的其他实施例，利用产品的结构参数构建的外观图还可以是三维立体结构图，用于立体地显示产品的形状、尺寸、材质以及外观设计等结构参数。所述外观相似度排序用于反映历史产品与产品在外观上的相似程度，所述外观可以由结构特征体现。
根据本公开实施例，所述对产品的外观图进行检测之前还可以对产品的外观图进行信号变换和降噪预处理，以去除信号中的杂质和干扰因素，从而提高后续识别的准确率。例如，所述信号变换可以包括将外观图的图像信号转换为电信号。
根据本公开实施例，将所述产品的外观图与历史产品的外观图进行比较得到外观相似度排序可以包括：对产品的外观图进行检测以提取产品的外观图中的结构特征，其中，对产品的外观图进行检测包括：利用扫描子窗口在产品的外观图中进行移动；对于扫描子窗口在移动过程中确定的产品的外观 图中的每个位置，计算该位置的结构特征。根据本公开实施例，将所述产品的外观图与历史产品的外观图进行比较得到外观相似度排序还可以包括：利用分类器对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序，以得到外观相似度排序。根据本公开实施例，可以利用分类器基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序。
根据本公开实施例，基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序可以包括以下步骤：将结构特征对应参数值转化为坐标值，得到初始特征排列；计算每个结构特征对应的权重，通过如下公式计算所述权重：
Figure PCTCN2019091122-appb-000002
其中，w为结构特征对应的权重，m为具有该结构特征的历史产品的总数，α i为权重比，(xi，yi)为用于表示图像结构特征的坐标值；基于每个结构特征对应的权重计算每个结构特征对应的排列分数，通过如下公式计算：c j＝(w j) 2；其中，w j为第j个结构特征对应的权重，c j为第j个结构特征对应的排列分数；在初始特征排列中去除排列分数最小的特征，并更新得到新的特征排列；重复上述循环过程直到所述特征排列中只包括一个特征，基于特征去除的顺序得到结构特征的相似度排序。
根据本公开实施例，所述计算机执行的报价方法还可以对所述分类器进行训练，所述训练可以包括利用训练样本数据、基于知识库或者限制条件训练所述分类器，其中，所述训练样本数据包括正样本和负样本，所述正样本为包含待检测结构特征的样本，所述负样本为不包含待检测结构特征的样本。
如图1A所示，接着，在步骤S103，将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序。其中，基于产品的外观图与历史产品的外观图的对比得到的外观相似度排序涉及产品的机械类零部件结构特征，除此之外，影响产品成本的关键因素还包括液晶面板、内部电路器件等。基于这些特征，可以建立对应于产品的电气参数的列表，用于进行产品的电气参数与历史产品的电气参数的相似度比较，以得到电气参数相似度排序，所述电气参数相似度排序用于反映历史产品与产品在电气参数上的 相似程度，排序越靠前，表明该历史产品在电气参数方面与产品越接近。
根据本公开实施例，可以首先确定历史产品具有的与产品相同的电气参数的数目，然后基于相同的电气参数的数目，对历史产品进行排序。举例来说，产品例如可以包括10个电气参数，通过比较(例如基于列表)，确定历史产品1具有8个与产品的电气参数相同的电气参数，历史产品2具有6个与产品的电气参数相同的电气参数，历史产品3具有5个与产品的电气参数相同的电气参数，从而可以基于相同的电气参数的数目，对历史产品进行排序，得到的排序诸如可以是历史产品1、历史产品2、历史产品3。
如图1A所示，接着，在步骤S104，基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序。外观相似度排序和电气参数相似度排序之后，需要基于外观相似度排序和电气参数相似度排序进行综合排序，即综合考虑历史产品与产品的外观相似度和电气参数相似度，以获得基于结构参数和电气参数的综合排序。可以根据结构部件(即机械零部件)和电气元件的成本权重计算所述基于结构参数和电气参数的综合排序。
根据本公开实施例，可以基于结构部件和电气元件的成本权重确定外观相似度排序和电气参数相似度的权重，基于所述外观相似度排序和电气参数相似度的权重计算得到基于结构参数和电气参数的综合排序。所述成本权重可以通过外观相似度排序以及电气参数相似度排序中，排序在前的多个历史产品对应的结构部件和电气元件所占成本权重求取平均值，确定外观相似度排序和电气参数相似度的权重，然后得到基于结构参数和电气参数的综合排序。
如图1A所示，在步骤S105，基于所述综合排序确定所述产品的物料清单，基于所述产品的物料清单计算所述产品报价。所述产品的物料清单(Bill of Material，BOM)是指产品所需的零部件明细表及其结构。作为一个示例，可以基于所述综合排序将排序最靠前的历史产品的物料清单作为所述产品的物料清单。对于所述历史产品，其物料清单可以与该历史产品的诸如电气参数和结构参数一起存储在计算机可访问的数据库内。在确定了综合排序最靠前的历史产品之后，计算机可以直接访问该数据库并提出所述历史产品的物理清单。其中，所述物料清单是构成装配件的所有子装配件、零件和原材料 的清单，也是制造一个装配件所需要每种零部件的数量的清单。此外，物料清单还可以与产品各个零件的单价进行关联。然后，基于所述产品的物料清单计算所述产品的报价。例如，基于产品的物料清单中包括的各种零件的数目以及单价确定所述产品的报价。
根据本公开实施例，确定所述产品的物料清单还可以包括：基于所述产品的结构参数和电气参数，对产品的物料清单进行调整，得到符合产品的结构参数和电气参数的产品的物料清单。也即，若是得到的产品物料清单中还是无法完全满足产品的所有参数要求，则需要相应的对物料清单进行一定程度的修改或者增删来使得物料清单中的零部件能够完全满足产品对于结构参数和电气参数的需求。然后，可以基于调整之后的产品的物料清单确定产品的报价。
根据本公开实施例，所述产品报价通过如下公式计算：
产品报价＝Σ(零部件成本)×加工费率×其他费率，
其中，零部件成本是基于所述产品的物料清单计算得到，加工费率和其他费率是从企业资源计划系统中得到。
例如，所述加工费率可以包含设备损耗、加工周期等，例如机器吨位、机器费用/小时、产品成型周期等。所述其他费率可以包含管理费用、运输费用以及包装费用等等。且加工费率和其他费率一般可以在企业资源计划系统(ERP)中获得。
图1B为本公开提供的报价方法的一个实施例的流程示意图，以下将结合图1B对根据本公开实施例的报价方法进行详细的描述。
如图1B所示，首先，在步骤S1，基于客户对产品的需求获得产品的结构参数和电气参数。接着，在步骤S2，利用产品的结构参数构建产品的外观图，并且将所述产品的外观图与历史产品的外观图进行相似度比较以得到外观相似度排序。
具体的，所述步骤S2可以包括步骤S21-S23，如图1B所示，在步骤S21，对产品的外观图进行检测并且基于产品的产品类型和结构特点相应提取得到产品外观图中的结构特征。例如，可以利用扫描子窗口的方式实现对于外观图的检测，具体包括：首先，利用扫描子窗口在待检测的产品的外观图中不断的进行移位滑动；其次，在移位滑动过程中产品的外观图中每个不同的位 置均通过扫描子窗口计算得到该位置区域的结构特征。所述对产品的外观图进行检测之前还可以对产品的外观图进行信号变换和降噪预处理。在，步骤S22，将提取得到的结构特征输入到预先构建的分类器中进行分类处理，得到各个结构特征的排序。其中，为了实现对于不同历史产品的排序，需要得到相应结构特征的排序，所述各个结构特征的排序指的是，对于各个结构特征中的每一个，确定历史产品与产品的该结构特征的排序，换句话说，在步骤S22可以获得基于每个结构特征的排序。
本公开中，可以采用特征递归消除算法计算得到结构特征的排序，具体步骤包括：
首先，将结构特征对应参数值转化为坐标值并且得到初始特征排列R＝[]。其中，R＝[]为一个特征数据形成的排列；其中R可能包含多个特征子集S＝{1,2,...,n}，也即特征排列中包括很多特征分类，然后每个特征分类中具有特征子集。
计算每个结构特征对应的权重；通过如下公式计算：
Figure PCTCN2019091122-appb-000003
其中，w为结构特征对应的权重；m为具有该结构特征的历史产品的总数；α i为权重比；(x i，y i)为用于表示图像结构特征的坐标值，例如可以用来表示图像特征。需要说明的是，这里将结构特征对应参数值转化为坐标值的方式可以选用当前任意可行的转化方案，本实施例不作具体限制。
基于每个结构特征对应的权重计算每个结构特征对应的排列分数；通过如下公式计算：
c j＝(w j) 2；其中，w j为第j个结构特征的权重；c j为第j个结构特征对应的排列分数。
在初始特征排列R＝[]中去除排列分数最小的特征并更新得到新的特征排列。具体的更新算法可以采用如下公式：e＝argmin(c)；R＝[e,R]；S＝S-[e]；也即每次排除一个排列分数最低的结构特征可以反向得到结构特征重要程度的排序；
重复上述循环过程直到特征排列只包括一个特征，即特征排列中剩下一 个特征，基于特征去除的顺序相应获取得到结构特征的排序。
图4为本公开提供的结构特征排序计算原理示意图。
需要说明的是，本申请上述实施例中采用的特征选择算法为基于支持向量机(Support Vector Machine)的特征递归消除算法(SVM-RFE)。
SVM-RFE是基于2范数SVM的，其数学模型为：
Figure PCTCN2019091122-appb-000004
s.t.    y i(w·x i+b)≥1-ξ i,i＝1,2,...,m
ξ i≥0,i＝1,2,...,m
其中，(xi，yi)是用空间中的一个坐标点来代表结构特征；C是一个常数；ξ i为误差；y i(w*x i+b)是函数间隔，用于实现约束条件的判断。w为结构特征对应的参数权向量，可以看出，通过变形，求解上式所描述的二次规划问题的过程就是求解一个凸优化问题。通过计算这个二次规划问题，可以得到特征的权向量w，每次移除特征后需要带入数据重新计算。在每次迭代中，特征的移除是基于SVM排列准则的，第j个有最小的排列分c j＝(w j) 2的特征将会被移除，w j是由SVM算出来的相应于第j个特征的权重。
这里，选择c j＝(w j) 2作为排列准则的原因是通过这种准则所移除的特征值将会对目标函数有最小的影响。在SVM-RFE中的目标函数是
Figure PCTCN2019091122-appb-000005
通过对目标函数的二阶泰勒级数的展开，可以估计由移除特征对目标函数造成的变化：
Figure PCTCN2019091122-appb-000006
一阶的导数可以忽略不计，代入
Figure PCTCN2019091122-appb-000007
则上述等式变成
ΔJ(j)＝(Δw j) 2
因此，该循环过程一直执行到特征集合中只剩一个特征，结果得到一列 按照特征重要性排序的结构特征序号列表。可以看出，SVM-RFE算法是一个后向搜索算法，在整个循环过程中，根据结构特征的排序分数，先是剔除了冗余的结构特征，然后再重新迭代计算得到一组新的排序分数，直到剩下对分类结果影响最大的结构特征组成的子集，因而可以起到减少结构特征维数，提高分类准确率的目的。
如图1B所示，在步骤S23，针对不同历史产品，基于产品结构特征的排序以及历史产品中相应的结构特征，综合计算历史产品基于结构特征的相似度排序，得到历史产品基于产品的外观相似度排序。
根据本公开实施例，所述计算历史产品基于结构特征的相似度排序的步骤还可以包括：
针对每个历史产品，相应的在产品结构特征的排序中提取得到当前历史产品所具有的结构特征以及相应的排序。通过历史产品的结构特征及排序相应的计算得到当前历史产品与产品的相似度。通过历史产品的结构特征以及当前产品结构特征的排序可以得到历史产品相对于产品结构特征的综合排序，也即得到所有历史产品与当前产品的外观相似度排序。
根据本公开实施例，可以利用分类器基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序。
图3为本公开提供的特征提取以及分类器训练的示意图。由图3可知，上述利用分类器实现分类之前还需要对分类器进行训练。所述分类器的训练样本数据包括正样本和负样本，其中，所述正样本为包含待检目标的样本，所述负样本为不包含待检目标的样本。此外，还可以基于知识库或者限制条件对分类器进行训练。由于样本数据的数据量较大，从而提取出来的特征的数据量也非常大，所以为了缩短对于分类器的训练的过程，可以加入知识库(例如，规则)或者引入限制条件来缩小搜索范围。例如通过知识库，可以判断一张电视机图片的某个位置代表按键的概率很大，这样就可以大大缩小搜索范围了。
如图1B所示，在步骤S3，将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序。接着，在步骤S4，基于历史产品中结构部件和电气元件所占成本权重的不同，通过外观相似度排序和电气参数相似度排序计算得到结构参数和电气参数对应历史产品的综合排序。
在步骤S5，基于综合排序在先的历史产品的物料清单，提取并构建得到产品的物料清单，相应计算得到产品的产品报价。在本申请一些可选的实施例中，所述提取得到产品的物料清单的步骤还包括：基于产品的结构参数和电气参数，对产品的物料清单进行调整和修正，用于使得产品的物料清单符合产品的结构参数和电气参数。也即，若是得到的产品物料清单中还是无法完全满足产品的所有参数要求，则需要相应的对物料清单进行一定程度的修改或者增删来使得物料清单中的零部件能够完全满足产品对于结构参数和电气参数的需求。
进一步的，所述产品报价通过如下公式计算：产品报价＝Σ(零部件成本)×加工费率×其他费率；产品报价在计算时需要综合考虑原料价格、加工费用和其他费用。其中，零部件成本主要为原材料价格，具体包括：原材料规格、原料价格、产品净重、胶口以及损耗等等。本申请所述计算机执行的报价方法中，通过产品的结构参数相应构建得到产品外观图，进而可以通过与历史产品外观图的识别和比较(诸如，通过分类器)得到外观相似度排序，同时利用电气参数比较得到历史产品的电气参数相似度排序，然后通过两个排序的综合分析得到历史产品的基于结构参数和电气参数的综合排序，这样可以直接将与产品较为相似的历史产品通过排序的方式提取出来，进而有利于后续利用历史产品的物料清单来相应的计算产品的报价，同时上述通过基于具体参数的比较得到的排序将会更加准确、计算效率更高，相对于人工筛选具有更好的准确率和报价速率。因此，本公开的方法不仅能够快速准确的实现产品的报价，而且大大降低了产品报价对于研发工程师的依赖，进一步降低了产品报价的人力物力成本。
图2为本公开提供的计算机执行的报价方法的示意性框图。由图2可知，根据本公开的报价方法可以包括两个主要部分，一是通过图像识别技术，将产品的外观图与历史产品的外观图进行比较得到外观相似度排序，例如，可以通过图像识别技术比较历史产品的外观图与产品的外观图。二是比较历史产品的电气参数和产品的电气参数得到电气参数相似度排序，基于外观相似度排序和电气参数相似度排序获得历史产品的综合相似度排序。接着，选择综合相似度排序中顺序将的历史产品，调出其BOM以及相应的价格信息，作为产品的物料清单。最后将产品BOM中各零部件的成本按照公式进行计 算，得出预估报价。
本公开提供的报价方法可以解决的问题包括：(1)可以解决目前产品报价不科学、不准确的问题；(2)打破销售与研发的部门壁垒，减少沟通时间，可以快速查询到近似产品作为报价参考；(3)可以通过科学的计算方法得到准确的产品报价；(4)帮助企业在获得最大利润的同时尽快实现成功竞标；(5)减少了人为查询历史产品价格信息的工作量。
本公开提供的报价方法、报价装置及电子设备，通过产品的结构参数构建得到产品的外观图，进而可以通过与历史产品的外观图进行相似度比较得到外观相似度排序，同时利用电气参数比较得到电气参数相似度排序，然后通过两个排序的综合分析得到历史产品的综合排序，从而提取出与产品较为相似的历史产品，以利用历史产品的物料清单来相应的计算产品的报价。上述通过基于具体参数的比较得到的排序将会更加准确、计算效率更高，相对于人工比较具有更高的准确率和更快的报价速率。因此，本申请不仅能够快速准确的实现产品的报价，而且大大降低了产品报价对于研发工程师的依赖，进一步降低了产品报价的人力物力成本。
参照图5所示，为本公开提供的产品报价装置的示意性结构框图。所述产品报价装置包括参数获取单元1、外观相似度排序单元2、电气参数相似度排序单元3、综合排序单元4以及产品报价单元5。
所述参数获取单元1可以配置成获取所述产品的结构参数和电气参数。并且参数获取单元1还可以配置成获取的将结构参数和电气参数发送到外观相似度排序单元2和电气参数相似度排序单元3中。
所述外观相似度排序单元2可以配置成利用产品的结构参数构建所述产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序。
所述电气参数相似度排序单元3可以配置成将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序。
所述综合排序单元4可以配置成基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序。
所述产品报价单元5可以配置成基于所述综合排序确定所述产品的物料 清单，并基于所述产品的物料清单计算所述产品的报价。
根据本公开的产品报价装置可以基于产品外观图和电气参数比较已有的历史产品，获得综合相似度排序较高的历史产品的物料清单，作为产品的物料清单，以计算新的产品报价。此外，还可以对历史产品的物料清单进行某些零部件的调整，将调整后的物料清单作为产品的物料清单，增加报价的准确性。
在根据本公开的一个实施例中，计算的产品报价可以发送到市场端，实现快速预估报价的效果，降低销售端与研发端的沟通难度及工作量，进一步的，还能够让客户快速获得产品的报价，从而加快了订单谈判的速度。
图6为本公开提供的执行报价方法的电子设备的硬件结构示意图。所述电子设备包括：至少一个处理器201以及存储器202，图6中以一个处理器201为例。所述存储器202存储有可被所述至少一个处理器执行的指令，使所述至少一个处理器能够执行如上所述的报价方法。
执行报价方法的电子设备还可以包括：输入装置203和输出装置204。处理器201、存储器202、输入装置203和输出装置204可以通过总线或者其他方式连接，图6中以通过总线连接为例。
存储器202作为一种非易失性计算机可读存储介质，可用于存储非易失性软件程序、非易失性计算机可执行程序以及模块，如本申请实施例中的报价方法对应的程序指令/模块。处理器201通过运行存储在存储器202中的非易失性软件程序、指令以及模块，从而执行服务器的各种功能应用以及数据处理，即实现上述方法实施例报价方法。
存储器202可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储根据产品报价装置的使用所创建的数据等。作为示例而非限定，所述存储器可以包括只读存储器(ROM)、随机访问存储器(RAM)，或者其他光盘存储、磁盘存储或其他磁存储设备、或能被用来承载或存储指令或数据结构形式的期望程序代码且能被计算机访问的任何存储介质。此外，存储器202可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实施例中，存储器202可选包括相对于处理器201远程设置的存储器，这些远程存储器可以通过网络 连接至产品报价装置。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
输入装置203可接收输入的数字或字符信息，以及产生与产品报价装置的用户设置以及功能控制有关的键信号输入。输出装置204可包括显示屏等显示设备。
所述一个或者多个模块存储在所述存储器202中，当被所述一个或者多个处理器201执行时，执行上述实施例中的报价方法。
本公开实施例的电子设备可以以多种形式存在，包括但不限于:
(1)移动通信设备：这类设备的特点是具备移动通信功能，并且以提供话音、数据通信为主要目标。这类终端包括:智能手机(例如iPhone)、多媒体手机、功能性手机，以及低端手机等。
(2)超移动个人计算机设备：这类设备属于个人计算机的范畴，有计算和处理功能，一般也具备移动上网特性。这类终端包括:PDA、MID和UMPC设备等，例如iPad。
(3)便携式娱乐设备：这类设备可以显示和播放多媒体内容。该类设备包括:音频、视频播放器(例如iPod)，掌上游戏机，电子书，以及智能玩具和便携式车载导航设备。
(4)服务器：提供计算服务的设备，服务器的构成包括处理器、硬盘、内存、系统总线等，服务器和通用的计算机架构类似，但是由于需要提供高可靠的服务，因此在处理能力、稳定性、可靠性、安全性、可扩展性、可管理性等方面要求较高。
(5)其他具有数据交互功能的电子装置。
根据本公开实施例的方法或装置也可以借助于图7所示的计算设备3000的架构来实现。如图7所示，计算设备3000可以包括总线3010、一个或多个CPU3020、只读存储器(ROM)3030、随机存取存储器(RAM)3040、连接到网络的通信端口3050、输入/输出组件3060、硬盘3070等。计算设备3000中的存储设备，例如ROM 3030或硬盘3070可以存储本公开提供的报价方法的处理和/或通信使用的各种数据或文件以及CPU所执行的程序指令。计算设备3000还可以包括用户界面3080。当然，图7所示的架构只是示例性的，在实现不同的设备时，根据实际需要，可以省略图7示出的计算设备 中的一个或多个组件。根据本公开实施例，还提供有一种计算机可读存储介质，其上存储有指令，所述指令在被处理器执行时，使得所述处理器执行如上所述的报价方法。
图8示出了根据本公开的存储介质的示意图4000。如图8所示，所述计算机存储介质4020上存储有计算机可读指令4010。当所述计算机可读指令4010由处理器运行时，可以执行参照以上附图描述的根据本公开实施例的报价方法。所述计算机可读存储介质包括但不限于例如易失性存储器和/或非易失性存储器。所述易失性存储器例如可以包括随机存取存储器(RAM)和/或高速缓冲存储器(cache)等。所述非易失性存储器例如可以包括只读存储器(ROM)、硬盘、闪存等。
所属领域的普通技术人员应当理解：以上任何实施例的讨论仅为示例性的，并非旨在暗示本公开的范围(包括权利要求)被限于这些例子；在本公开的思路下，以上实施例或者不同实施例中的技术特征之间也可以进行组合，步骤可以以任意顺序实现，并存在如上所述的本公开的不同方面的许多其它变化，为了简明它们没有在细节中提供。
另外，为简化说明和讨论，并且为了不会使本公开难以理解，在所提供的附图中可以示出或可以不示出与集成电路(IC)芯片和其它部件的公知的电源/接地连接。此外，可以以框图的形式示出装置，以便避免使本公开难以理解，并且这也考虑了以下事实，即关于这些框图装置的实施方式的细节是高度取决于将要实施本公开的平台的(即，这些细节应当完全处于本领域技术人员的理解范围内)。在阐述了具体细节(例如，电路)以描述本公开的示例性实施例的情况下，对本领域技术人员来说显而易见的是，可以在没有这些具体细节的情况下或者这些具体细节有变化的情况下实施本公开。因此，这些描述应被认为是说明性的而不是限制性的。
尽管已经结合了本公开的具体实施例对本公开进行了描述，但是根据前面的描述，这些实施例的很多替换、修改和变型对本领域普通技术人员来说将是显而易见的。例如，其它存储器架构(例如，动态RAM(DRAM))可以使用所讨论的实施例。
本公开的实施例旨在涵盖落入所附权利要求的宽泛范围之内的所有这样的替换、修改和变型。因此，凡在本公开的精神和原则之内，所做的任何省 略、修改、等同替换、改进等，均应包含在本公开的保护范围之内。
本申请要求于2018年07月05日提交的中国专利申请第201810731140.4号的优先权，该中国专利申请的全文通过引用的方式结合于此以作为本申请的一部分。

Claims (15)

  1. 一种计算机执行的报价方法，包括：
    获取产品的结构参数和电气参数；
    利用产品的结构参数构建所述产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序；
    将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序；
    基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序；以及
    基于所述综合排序确定所述产品的物料清单，基于所述产品的物料清单计算所述产品报价。
  2. 根据权利要求1所述的方法，其中，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序包括：
    对产品的外观图进行检测以提取产品的外观图中的结构特征；
    利用分类器对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序，以得到外观相似度排序。
  3. 根据权利要求2所述的方法，还包括：对产品的外观图进行信号变换和降噪预处理。
  4. 根据权利要求2所述的方法，其中，对产品的外观图进行检测包括：
    利用扫描子窗口在产品的外观图中进行移动；
    对于扫描子窗口在移动过程中确定的产品的外观图中的每个位置，计算该位置的结构特征。
  5. 根据权利要求2所述的方法，其中，利用分类器对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序包括：
    利用分类器基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序。
  6. 根据权利要求5所述的方法，其中，基于特征递归消除算法对历史产品的外观图中的结构特征与产品的外观图中的结构特征进行相似度排序包括：
    将结构特征对应参数值转化为坐标值，得到初始特征排列；
    计算每个结构特征对应的权重，通过如下公式计算所述权重：
    Figure PCTCN2019091122-appb-100001
    其中，w为结构特征对应的权重，m为具有该结构特征的历史产品的总数，α i为权重比，(x i，y i)为用于表示图像结构特征的坐标值；
    基于每个结构特征对应的权重计算每个结构特征对应的排列分数，通过如下公式计算：
    c j＝(w j) 2；其中，w j为第j个结构特征对应的权重，c j为第j个结构特征对应的排列分数；
    在初始特征排列中去除排列分数最小的特征，并更新得到新的特征排列；
    重复上述循环过程直到所述特征排列中只包括一个特征，基于特征去除的顺序得到结构特征的相似度排序。
  7. 根据权利要求2所述的方法，还包括利用训练样本数据、基于知识库或者限制条件训练所述分类器，其中，所述训练样本数据包括正样本和负样本，所述正样本为包含待检测结构特征的样本，所述负样本为不包含待检测结构特征的样本。
  8. 根据权利要求1所述的方法，其中，所述外观图中包括产品的形状、尺寸、材质以及外观设计参数，且所述外观图为采用同一比例绘制的产品六视图。
  9. 根据权利要求1所述的方法，其中，将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序包括：
    确定历史产品具有的与产品相同的电气参数的数目；
    基于相同的电气参数的数目，对历史产品进行排序。
  10. 根据权利要求1所述的方法，其中，基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序计算得到基于结构参数和电气参数的综合排序包括：
    基于结构部件和电气元件的成本权重确定外观相似度排序和电气参数相似度的权重；
    基于所述外观相似度排序和电气参数相似度的权重计算得到基于结构参数和电气参数的综合排序。
  11. 根据权利要求1所述的方法，其中，确定所述产品的物料清单还包括：基于所述产品的结构参数和电气参数，对产品的物料清单进行调整，得到符合产品的结构参数和电气参数的产品的物料清单。
  12. 根据权利要求1所述的方法，其中，所述产品报价通过如下公式计算：
    产品报价＝Σ(零部件成本)×加工费率×其他费率；
    其中，零部件成本是基于所述产品的物料清单计算得到，加工费率和其他费率是从企业资源计划系统中得到。
  13. 一种报价装置，包括：
    参数获取单元，配置成获取所述产品的结构参数和电气参数；
    外观相似度排序单元，配置成利用产品的结构参数构建所述产品的外观图，将所述产品的外观图与历史产品的外观图进行相似度比较得到外观相似度排序；
    电气参数相似度排序单元，配置成将产品的电气参数与历史产品的电气参数进行相似度比较得到电气参数相似度排序；
    综合排序单元，配置成基于结构部件和电气元件的成本权重以及外观相似度排序和电气参数相似度排序得到基于结构参数和电气参数的综合排序；以及
    产品报价单元，配置成基于所述综合排序确定所述产品的物料清单，并基于所述产品的物料清单计算所述产品报价。
  14. 一种电子设备，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，使所述至少一个处理器能够执行如权利要求1-12中任一项所述的计算机执行的报价方法。
  15. 一种计算机可读存储介质，其上存储有指令，所述指令在被处理器执行时，使得所述处理器执行如权利要求1-12中任一项所述的计算机执行的报价方法。
PCT/CN2019/091122 2018-07-05 2019-06-13 计算机执行的报价方法、报价装置、电子设备及存储介质 WO2020007177A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2020568293A JP7405775B2 (ja) 2018-07-05 2019-06-13 コンピュータで実行される見積もり方法、見積もり装置、電子機器及び記憶媒体
US16/633,214 US11544751B2 (en) 2018-07-05 2019-06-13 Quotation method executed by computer, quotation device, electronic device and storage medium
EP19830747.2A EP3819854A4 (en) 2018-07-05 2019-06-13 PRICE OFFERING PROCESS EXECUTED BY A COMPUTER, PRICE OFFERING DEVICE, ELECTRONIC DEVICE AND STORAGE MEDIA

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810731140.4A CN109064253A (zh) 2018-07-05 2018-07-05 一种新产品报价方法、报价装置及电子设备
CN201810731140.4 2018-07-05

Publications (1)

Publication Number Publication Date
WO2020007177A1 true WO2020007177A1 (zh) 2020-01-09

Family

ID=64819377

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/091122 WO2020007177A1 (zh) 2018-07-05 2019-06-13 计算机执行的报价方法、报价装置、电子设备及存储介质

Country Status (5)

Country Link
US (1) US11544751B2 (zh)
EP (1) EP3819854A4 (zh)
JP (1) JP7405775B2 (zh)
CN (1) CN109064253A (zh)
WO (1) WO2020007177A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109064253A (zh) 2018-07-05 2018-12-21 京东方科技集团股份有限公司 一种新产品报价方法、报价装置及电子设备
US12008104B2 (en) * 2020-03-31 2024-06-11 Onapsis Inc. Transport security in business applications
CN112182052A (zh) * 2020-10-12 2021-01-05 中国铁塔股份有限公司 一种参数展示方法、装置、电子设备和可读存储介质
CN112581221A (zh) * 2020-12-10 2021-03-30 北京索为云网科技有限公司 工业产品配置及报价方法及系统
CN113537794A (zh) * 2021-07-22 2021-10-22 北京中科闻歌科技股份有限公司 对目标对象的分析方法、装置、电子设备及存储介质
CN115688483B (zh) * 2022-11-30 2023-03-21 雪峰创新(北京)科技有限公司 适用于定制化制造的数据智能处理系统
CN116523497B (zh) * 2023-04-07 2023-11-17 国网安徽省电力有限公司滁州供电公司 基于pms集成工单运维成本量化系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014055907A1 (en) * 2012-10-04 2014-04-10 Valuerz, Inc. Systems and methods for determining a fair price range for commodities
CN105956875A (zh) * 2016-04-21 2016-09-21 中国农业大学 基于大数据与市价匹配的农业科技成果估价方法及系统
CN108182606A (zh) * 2018-01-25 2018-06-19 北京理工大学 一种大规模定制产品快速成本估算方法
CN108198072A (zh) * 2017-12-29 2018-06-22 郑州云海信息技术有限公司 一种人工智能评估金融产品特征的系统
CN109064253A (zh) * 2018-07-05 2018-12-21 京东方科技集团股份有限公司 一种新产品报价方法、报价装置及电子设备

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004164274A (ja) * 2002-11-13 2004-06-10 Topy Ind Ltd 見積金額決定システム及び方法
CN1661606A (zh) * 2004-02-23 2005-08-31 鸿富锦精密工业（深圳）有限公司 产品报价系统及方法
JP5127358B2 (ja) * 2007-08-13 2013-01-23 旭化成ホームズ株式会社 住宅設備機器選定支援システム
CN103246950B (zh) * 2012-10-30 2016-08-03 中国科学院沈阳自动化研究所 一种半导体封装测试企业的订单承诺方法
US9741018B2 (en) * 2014-10-28 2017-08-22 Ebay Inc. Systems and methods for extracting similar group elements
CN106126755A (zh) * 2016-08-24 2016-11-16 广东华邦云计算股份有限公司 一种基于图像识别的购物方法
JP2018073087A (ja) * 2016-10-28 2018-05-10 株式会社日立製作所 デバイス設計受付システム
JP6831676B2 (ja) * 2016-11-15 2021-02-17 株式会社カブク 価格見積もりシステム、価格見積もり方法及びプログラム

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014055907A1 (en) * 2012-10-04 2014-04-10 Valuerz, Inc. Systems and methods for determining a fair price range for commodities
CN105956875A (zh) * 2016-04-21 2016-09-21 中国农业大学 基于大数据与市价匹配的农业科技成果估价方法及系统
CN108198072A (zh) * 2017-12-29 2018-06-22 郑州云海信息技术有限公司 一种人工智能评估金融产品特征的系统
CN108182606A (zh) * 2018-01-25 2018-06-19 北京理工大学 一种大规模定制产品快速成本估算方法
CN109064253A (zh) * 2018-07-05 2018-12-21 京东方科技集团股份有限公司 一种新产品报价方法、报价装置及电子设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3819854A4

Also Published As

Publication number Publication date
EP3819854A1 (en) 2021-05-12
JP2021529369A (ja) 2021-10-28
US11544751B2 (en) 2023-01-03
EP3819854A4 (en) 2022-03-16
JP7405775B2 (ja) 2023-12-26
CN109064253A (zh) 2018-12-21
US20200193491A1 (en) 2020-06-18

Similar Documents

Publication Publication Date Title
WO2020007177A1 (zh) 计算机执行的报价方法、报价装置、电子设备及存储介质
US20220198289A1 (en) Recommendation model training method, selection probability prediction method, and apparatus
US10394854B2 (en) Inferring entity attribute values
EP3028184B1 (en) Method and system for searching images
WO2021143267A1 (zh) 基于图像检测的细粒度分类模型处理方法、及其相关设备
CN112184508B (zh) 一种用于图像处理的学生模型的训练方法及装置
US9875294B2 (en) Method and apparatus for classifying object based on social networking service, and storage medium
US11126660B1 (en) High dimensional time series forecasting
WO2021135562A1 (zh) 特征有效性评估方法、装置、电子设备及存储介质
EP4322031A1 (en) Recommendation method, recommendation model training method, and related product
WO2020233432A1 (zh) 一种信息推荐方法及装置
CN106610970A (zh) 基于协同过滤的内容推荐系统与方法
WO2021031410A1 (zh) 基于商品信息预测模型的商品信息推送方法及装置、非易失性可读存储介质、计算机设备
CN112395487A (zh) 信息推荐方法、装置、计算机可读存储介质及电子设备
CN111966886A (zh) 对象推荐方法、对象推荐装置、电子设备及存储介质
WO2024041483A1 (zh) 一种推荐方法及相关装置
WO2023185925A1 (zh) 一种数据处理方法及相关装置
Naseri et al. Novel CNN with investigation on accuracy by modifying stride, padding, kernel size and filter numbers
WO2020135054A1 (zh) 视频推荐方法、装置、设备及存储介质
CN111210022A (zh) 向后模型选择方法、设备及可读存储介质
CN102760127B (zh) 基于扩展文本信息来确定资源类型的方法、装置及设备
CN117036897A (zh) 一种基于Meta RCNN的少样本目标检测方法
CN111275683A (zh) 图像质量评分处理方法、系统、设备及介质
CN112650869B (zh) 图像检索重排序方法、装置、电子设备及存储介质
US20240135576A1 (en) Three-Dimensional Object Detection

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19830747

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020568293

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2019830747

Country of ref document: EP