Engine inspection report intelligent generation method and system and medium

Abstract

The invention discloses an engine inspection report intelligent generation method and system and a medium, and the method comprises the steps: obtaining fault pictures, importing the fault pictures into a fault information table in batches, calling and traversing all to-be-replaced parts in an inspection report template, and when a tag type in the to-be-replaced parts is traversed to be characters and a tag name is the same as a field name in the fault information table, executing the step 1; filling corresponding field values in the fault information table into the to-be-filled text labels; when the label type in the to-be-replaced part is traversed to be a table, classification is performed according to classification field values with classification field names in the fault information table, a plurality of pieces of data corresponding to the same classification field values are filled into the same part information table template, and finally a plurality of part information tables are obtained; filling the to-be-filled table labels with a plurality of part information tables; after all the to-be-replaced parts of the inspection report template are traversed, an engine inspection report is output. And the inspection report is automatically generated and output according to the inspection report template, so that the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of engine inspection report data processing, and in particular to an engine inspection report intelligent generation method, system and medium. Background Art

[0002] With the rapid development of the aviation industry, the importance of engine maintenance and inspection work has become increasingly prominent. At present, the generation of engine maintenance and inspection reports mainly relies on manual operations. After the engine maintenance and inspection work is completed, when the inspection data is sorted out, the information is generally entered manually. It is necessary to manually copy the data from the original inspection data, and then integrate the relevant content in the target report template (such as text, tables, pictures and other information), and manually calculate, analyze, and layout to obtain a complete inspection report. However, in this way, when manually entering information such as the engine serial number and part number, input errors are prone to occur, and the accuracy of the basic information needs to be repeatedly checked, which is labor-intensive and inefficient; and when manually entering, the information format is not unified. For example, when inserting pictures in the inspection report, the size of each picture needs to be manually adjusted, and the pictures need to be manually placed in the corresponding position. The storage and management of pictures are chaotic, retrieval is difficult, and the difficulty of post-processing is increased. Summary of the invention

[0003] The purpose of the present invention is to solve the problems of low automation and low efficiency in the existing inspection report generation process, and to provide an engine inspection report intelligent generation method that can automatically generate an inspection report output according to an inspection report template, thereby improving work efficiency.

[0004] In order to achieve the above objectives, this application proposes the following solutions: In one aspect, the present invention provides a method for intelligently generating an engine inspection report, which specifically comprises the following steps: S1. Obtain fault pictures in batches and import them into the fault information table, and fill the fault information table according to the parts information in the fault pictures; S2. Calling the inspection report template, traversing each part to be replaced in the inspection report template, each part to be replaced includes a plurality of labels to be filled, and the type of the label to be filled includes text or table; S21, when the tag type in the part to be replaced is text and the tag name is the same as the field name in the fault information table, fill the corresponding field value in the fault information table into the text tag to be filled; S22, when the tag type in the part to be replaced is a table, classification is performed according to the classification field value of the field name classification in the fault information table, and each row of data in the fault information table is taken as a piece of data, and finally a plurality of pieces of data corresponding to the same classification field value are obtained; Call several parts information table templates, fill several pieces of data corresponding to the same classification field value into the same parts information table template, and finally obtain several parts information tables, and fill the several parts information tables into the table labels to be filled; S3. After traversing all the parts to be replaced in the inspection report template, output the engine inspection report.

[0005] In some specific implementation schemes, the format of the inspection report template is Word, and the parts to be replaced include the Word header, title and body.

[0006] In some specific implementation schemes, each field in the fault information table is a column, and each row includes field values ​​of multiple fields, and the fields include classification, part number, part name, fault picture and fault description, wherein the classification field values ​​include time-life parts, power section, gearbox section, and other parts.

[0007] In some specific implementation schemes, the specific process of filling several pieces of data corresponding to the same classification field value into the same part information table template is as follows: S01. The part information table template includes a title module and an adjustable module. The title module includes a text label to be filled in, and the current classification field value replaces the text label to be filled in; S02, according to the n pieces of data corresponding to the current classification field value, the number of areas to be filled in the adjustable module in the part information table template is set to n, and each area to be filled in includes a picture label and a text label to be filled; S03, searching for the corresponding label type to be filled in the area to be filled according to the field name for each piece of data and replacing the corresponding label to be filled; S04. Adaptively create a table, layout and display the n areas to be filled in the table, and finally obtain a part information table corresponding to the current classification field value.

[0008] In some specific implementation schemes, the specific process of adaptively creating a table and typeset displaying n areas to be filled in the table is as follows: The display area size of each area to be filled in is set to the same size, and the n areas to be filled in are arranged and displayed in a row of two at a time. The final size of the table for the layout display of the adjustable module is: 2×[n / 2], where [n / 2] is rounded up. When n is an even number, the table size is n / 2 rows and 2 columns. When n is an odd number, the table size is (n+1) / 2 rows and 2 columns.

[0009] In some specific implementation schemes, when n pieces of data are filled into the area to be filled in and laid out and displayed in a table, the n pieces of data are screened again according to the part number, and several pieces of data with the same part number are laid out and displayed continuously. When there are two pieces of data with the same part number and the corresponding fault pictures include an overall picture of the part and a partial picture of the fault, the two pieces of data containing the overall picture of the part and the partial picture of the fault are displayed side by side in the same row, and an identification symbol is added to point from the overall picture of the part to the partial picture of the fault.

[0010] In some specific implementation schemes, the text label includes a first text label and a second text label, and when each piece of data is filled into each area to be filled, when the image label to be filled is replaced with the image of the fault image field, the image of the fault image field is adjusted to a uniform preset size and then filled into the image label position; When the field value of the fault description field is used to replace the position corresponding to the fault description in the first text label, the preset professional corpus is called to translate the field value of the fault description field, and the translated content replaces the position corresponding to the fault description in the second text label.

[0011] In some specific implementation schemes, the method of filling the fault information table in step S1 is: Import the fault picture into the picture field of the information table, call the part classification model to identify the parts contained in the picture, find the corresponding classification and part number in the parts library according to the part name, and automatically fill the recognition result into the part name, classification field and part number field. Call the part fault recognition model to automatically identify the fault type in the fault picture and mark the fault location in the fault picture, generate a fault description and fill it into the fault description field and refresh the fault picture field with the overall picture of the part with annotations. At the same time, extract the fault local picture from the overall picture of the part with annotations and insert it into the table below the fault picture, and copy the values ​​of other fields in the row where the overall picture of the part with annotations is located.

[0012] In a second aspect, the present application provides an engine inspection report intelligent generation system, comprising: Fault preprocessing module, used to obtain fault pictures and import them into the fault information table in batches, and fill in the fault information according to the parts information in the fault pictures; The inspection report analysis module is used to call the inspection report template and traverse each part to be replaced in the inspection report template, each part to be replaced includes a plurality of labels to be filled, and the types of labels to be filled include text or table; The inspection report generation module is used to fill the corresponding field value in the fault information table into the text label to be filled when the label type in the part to be replaced is text and the label name is the same as the field name in the fault information table; When the tag type in the part to be replaced is a table, classification is performed according to the classification field value of the field name classification in the fault information table, and each row of data in the fault information table is taken as a piece of data, and finally several pieces of data corresponding to the same classification field value are obtained; Call several parts information table templates, fill several pieces of data corresponding to the same classification field value into the same parts information table template, and finally obtain several parts information tables, and fill the several parts information tables into the table labels to be filled; The inspection report output module is used to output the engine inspection report after traversing all the parts to be replaced in the inspection report template.

[0013] In a third aspect, the present application provides a computer-readable storage medium, comprising: one or more processors; A storage unit is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors can implement the method for intelligently generating an engine inspection report as described in the first aspect.

[0014] The present invention has the beneficial effects: The present invention realizes intelligent, automated and standardized generation of engine inspection reports through automatic information extraction, automatic image processing and automatic report generation without manual input and manual adjustment of image display, thereby improving the efficiency of report generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of an engine inspection report intelligent generation method provided by an embodiment of the present invention; Figure 2 A schematic diagram of the fault information provided by the embodiment of the present invention; Figure 3 A schematic diagram of a parts information table template provided in an embodiment of the present invention; Figure 4 A schematic diagram of a parts information table provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0018] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0019] Additionally, descriptions of well-known structures, functions, and configurations may be omitted for clarity and conciseness.One of ordinary skill in the art will recognize that various changes and modifications may be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0020] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0021] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0022] Example 1 like Figure 1 As shown, this embodiment provides an intelligent generation method for an engine inspection report, which specifically includes the following steps: S1. Obtain fault pictures and import them into the fault information table in batches, and fill in the fault information table according to the parts information in the fault pictures; the fault information table is used to record the information of each part of the engine, and the table name or title of the fault information table can be named and filled with relevant information of the engine. Each field in the fault information table is a column, and each row includes field values ​​of multiple fields, and the fields include classification, part number, part name, fault picture and fault description, among which the classification field values ​​include time-life parts, power section, gearbox section, and other parts.

[0023] Specifically, the fault picture is a whole picture of the engine parts taken by the staff. In order to realize automatic report generation, the whole picture of the parts is classified and the fault is identified. Specifically, the method of filling the fault information table in step S1 is: Import the fault image into the image field of the information table, call the part classification model to identify the parts contained in the image, search for the corresponding classification and part number in the parts library according to the part name, and automatically fill the recognition result into the part name, classification field and part number field; Call the part fault recognition model to automatically identify the fault type in the fault image and mark the fault location in the fault image, generate a fault description and fill it into the fault description field and refresh the fault image field with the overall picture of the part with annotations. At the same time, extract the fault local picture from the overall picture of the part with annotations and insert it into the table below the fault picture, and copy the values ​​of other fields in the row where the overall picture of the part with annotations is located.

[0024] like Figure 2 As shown, it can be seen that the fault picture field imports pictures of various engine parts in sequence. In order to facilitate the subsequent automatic filling of the inspection report template, the fault information table is first filled in. The fields to be filled in are part name, part number, part classification and fault description. Here, the fault picture can be identified by the pre-trained part classification model to obtain the part name, and then the part name can be automatically searched in the parts library and filled in the table according to the corresponding fields. Further, the fault identification of the overall part picture can be performed by the pre-trained model, and the fault picture can also be manually annotated. In order to more intuitively show the fault location to the user, when a defect is found in the overall part picture through fault identification, the defect is identified as the fault area and marked in the overall part picture in a box selection manner. In order to more intuitively show the fault location, the fault part is cut out from the overall part picture, the fault type is identified, and the fault description is obtained and filled in the fault description field. At the same time, the fault part picture corresponding to the fault part is cut out and placed below the corresponding overall part picture. In addition, the two data of the overall part picture and the fault part picture in the fault picture field of the same part and the same part number are different except for the fault picture field value. The other field values ​​are the same. And the two pieces of data of the overall picture of the part and the partial picture of the fault are bound and fixed as a group. When generating the inspection report template, the two pieces of data can be read in sequence in the order of the overall picture of the part and the partial picture of the fault for continuous display.

[0025] S2. Calling the inspection report template, traversing each part to be replaced in the inspection report template, each part to be replaced includes a plurality of labels to be filled, and the type of the label to be filled includes text or table; The format of the inspection report template is Word, and the parts to be replaced include the Word header, title, and body. Generally, there are texts that need to be replaced in the Word header and title, such as basic information such as engine model, serial number, part number, working / cycle duration, etc. For the text that needs to be replaced, a string in a specific format is generally used as a label to mark the position. In this way, when the computer traverses the inspection report template, when it traverses to this string in a specific format, it can find the corresponding field name that needs to be replaced in the fault information table according to the name of the string, and then replace it. For example, the engine name may need to be replaced in the Word header. At this time, the text label may be #{engine model}#. Find the information corresponding to "engine model" in the fault information table and replace #{engine model}#. In the body part, all pictures and text need to be embedded in the report template. When embedding pictures, the picture size and layout are automatically laid out for more intuitive display. In addition, the pictures need to be numbered when displayed. The picture numbers correspond to the numbers of the pictures in the fault information table. Therefore, in the body part: S21, when the tag type in the part to be replaced is text and the tag name is the same as the field name in the fault information table, fill the corresponding field value in the fault information table into the text tag to be filled; S22, when the tag type in the part to be replaced is a table, classification is performed according to the classification field value of the field name classification in the fault information table, and each row of data in the fault information table is taken as a piece of data, and finally a plurality of pieces of data corresponding to the same classification field value are obtained; Call several parts information table templates, traverse the fault information table, fill several pieces of data corresponding to the same classification field value into the same parts information table template, and finally obtain several parts information tables, and fill the several parts information tables into the table tags to be filled; Specifically, in order to realize the adaptive layout of pictures in the table, the specific process of filling several pieces of data corresponding to the same classification field value into the same part information table template is as follows: S01. The part information table template includes a title module and an adjustable module. The title module includes a text label to be filled in, and the current classification field value replaces the text label to be filled in; S02, according to the n pieces of data corresponding to the current classification field value, the number of areas to be filled in the adjustable module in the part information table template is set to n, and each area to be filled in includes a picture label and a text label to be filled; The text label of each area to be filled in includes a first text label and a second text label. When filling each piece of data into each area to be filled in, when using the picture of the fault picture field to replace the picture label to be filled, the picture of the fault picture field is adjusted to a uniform preset size and then filled into the picture label position; when displayed, each piece of data is filled in sequentially with the picture on top and the text on the bottom, with a unified paragraph format and picture size.

[0026] When the field value of the fault description field is used to replace the position corresponding to the fault description in the first text label, the preset professional corpus is called to translate the field value of the fault description field, and the translated content replaces the position corresponding to the fault description in the second text label. In addition, in order to facilitate other personnel to view and record, the field value of the fault description field in the general fault information table is described in Chinese, while the existing report template generally uses English descriptions. In order to describe in accordance with professionalism and standards, a corpus of engine terms is pre-established. Therefore, in order to facilitate staff to view and comply with specifications, when filling in the report template, the Chinese description is translated according to the corpus to obtain other language descriptions that conform to the report template (such as English descriptions), and finally displayed in bilingual form in the report template.

[0027] S03, searching for the corresponding label type to be filled in the area to be filled according to the field name for each piece of data and replacing the corresponding label to be filled; In order to better fill n pieces of data into the area to be filled in and display them in the table, the n pieces of data are screened again according to the part number. If there are multiple pieces of data with the same part number, the pieces of data with the same part number are sorted as a group and displayed continuously in priority. For example, when there are two pieces of data with the same part number and the corresponding fault pictures include the overall picture of the part and the partial picture of the fault, the pictures of the two pieces of data containing the overall picture of the part and the partial picture of the fault are bound. For the text label in the area to be filled in, such as #{fault description}#, the corresponding field is found from the fault information table, and the field value corresponding to the piece of data is read and filled in. For the picture label such as ##{fault picture}##, the picture of the corresponding fault picture field is adjusted to a preset fixed size and replaced to the position of the fault picture, and then it is determined whether the overall picture of the part is bound to the corresponding partial picture of the fault. If so, the corresponding partial picture of the fault is set in the picture label ##{fault picture}## in another area to be filled in the same row as the overall picture of the part for display, and an identification symbol is added to point from the overall picture of the part to the partial picture of the fault. Figure 4 shown.

[0028] S04. Adaptively create a table, layout and display the n areas to be filled in the table, and finally obtain a part information table corresponding to the current classification field value.

[0029] Specifically, the specific process of adaptively creating a table and typeset displaying n areas to be filled in the table is as follows: The display area size of each area to be filled is set to the same size, and n areas to be filled are arranged and displayed in rows of two each. The size of the table for the layout display of the adjustable module is finally obtained as follows: 2×[n / 2], where [n / 2] is rounded up. When n is an even number, the table size is n / 2 rows and 2 columns, and when n is an odd number, the table size is (n+1) / 2 rows and 2 columns. For example, Figure 3 As shown in the figure, when there are three pieces of data, there are only three areas to be filled in, but the size of the created table needs to be displayed in a 2×2 manner. At this time, the layout display of the areas to be filled in is to fill the table in order from left to right and from top to bottom. Each table uses a field value of a classification field as the table name, such as Figure 3 The figure shows that there are three areas to be filled in, and the labels to be filled in each area to be filled in are arranged in the manner shown in the figure. The first text label includes: Picture 1: #{part name}# #{part number}# #{fault description}#; The second text label includes: Image 1: #{Part Name}# #{Part Number}# #{Fault Description}# The difference between the first text label and the second text label is that the #{fault description}# of the first text label directly obtains the corresponding field value from the fault description field in the fault information table for replacement, while the #{fault description}# of the second text label needs to be replaced by the content of the field after the first text label replaces #{fault description}# in the corpus and the output content after translation. Others such as part name and part number can be directly found in the corresponding field in the fault information table, and the field value corresponding to the data can be read and filled in. The part information table obtained after replacing the corresponding label with the field value is as follows: Figure 4 shown.

[0030] For example, according to Figure 2 The fault information table shown in the figure summarizes the parts classified as Life Limit Parts and named Turbine Rotor and numbers them 1, 2, and 3 respectively. Figure 4 In the picture 1, picture 2 and picture 3, when the part number 3840310-4 has a crack fault, fill in the corresponding overall picture and the faulty local picture in the fault information table. Figure 2 The three data in the fault information table are generated as follows: Figure 3As shown in the parts information table template, fill in the classification Life Limit Parts corresponding to the three data into #{classification}#, and fill in the three data into each area to be filled in turn. Taking the first area to be filled in as an example, replace the picture corresponding to the fault picture field of the first data numbered 1 in the fault information table with ##{fault picture}##. For each label to be replaced in the first text label, replace #{part name}# with "Turbine Rotor", replace #{part number}# with "3840310-4", and replace #{fault description}# with "cracks have appeared". Then copy the "TurbineRotor" and "3840310-4" of the first text label to replace #{part name}# and #{part number}# of the second text label, and translate "cracks have appeared" of the first text label into "Cracks have appeared" and replace it in #{fault description}# of the second text label.

[0031] S3. After traversing all the parts to be replaced in the inspection report template, output the engine inspection report.

[0032] Before outputting the engine inspection report, after traversing all the parts to be replaced in the inspection report template, the inspection report to be summarized is obtained, and then the entire inspection report to be summarized can be traversed by calling the big language AI model, and then the preset prompt words are input into the big language AI model to summarize the inspection report to be summarized, for example, how long the engine has been used, the main cause of the failure, etc. Finally, the report summary is output and inserted into the end part of the inspection report to be summarized to obtain the engine inspection report.

[0033] Example 2 In order to apply the automatic generation method of Example 1, this embodiment provides an engine inspection report intelligent generation system, including: Fault preprocessing module, used to obtain fault pictures and import them into the fault information table in batches, and fill in the fault information according to the parts information in the fault pictures; The inspection report analysis module is used to call the inspection report template and traverse each part to be replaced in the inspection report template, each part to be replaced includes a plurality of labels to be filled, and the types of labels to be filled include text or table; The inspection report generation module is used to fill the corresponding field value in the fault information table into the text label to be filled when the label type in the part to be replaced is text and the label name is the same as the field name in the fault information table; When the tag type in the part to be replaced is a table, classification is performed according to the classification field value of the field name classification in the fault information table, and each row of data in the fault information table is taken as a piece of data, and finally several pieces of data corresponding to the same classification field value are obtained; Call several parts information table templates, fill several pieces of data corresponding to the same classification field value into the same parts information table template, and finally obtain several parts information tables, and fill the several parts information tables into the table labels to be filled; The inspection report output module is used to output the engine inspection report after traversing all the parts to be replaced in the inspection report template.

[0034] Example 3 This embodiment provides a computer-readable storage medium, including: one or more processors; A storage unit is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors can implement the method for intelligently generating an engine inspection report described in Example 1.

[0035] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. According to the technical essence of the present invention, within the spirit and principles of the present invention, any simple modification, equivalent replacement and improvement made to the above embodiment still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for intelligently generating an engine inspection report, characterized in that: The specific steps include: S1. Obtain fault pictures in batches and import them into the fault information table, and fill the fault information table according to the parts information in the fault pictures; S2. Calling the inspection report template, traversing each part to be replaced in the inspection report template, each part to be replaced includes a plurality of labels to be filled, and the type of the label to be filled includes text or table; S21, when the tag type in the part to be replaced is text and the tag name is the same as the field name in the fault information table, fill the corresponding field value in the fault information table into the text tag to be filled; S22, when the tag type in the part to be replaced is a table, classification is performed according to the classification field value of the field name classification in the fault information table, and each row of data in the fault information table is taken as a piece of data, and finally a plurality of pieces of data corresponding to the same classification field value are obtained; Call several parts information table templates, fill several pieces of data corresponding to the same classification field value into the same parts information table template, and finally obtain several parts information tables, and fill the several parts information tables into the table labels to be filled; S3. After traversing all the parts to be replaced in the inspection report template, output the engine inspection report.

2. The method for intelligently generating an engine inspection report according to claim 1, characterized in that: The format of the inspection report template is Word, and the parts to be replaced include the Word header, title and body.

3. The method for intelligently generating an engine inspection report according to claim 1, characterized in that: Each field in the fault information table is a column, and each row includes field values ​​of multiple fields. The fields include classification, part number, part name, fault picture and fault description. Among them, the classification field values ​​include time-limited parts, power section, gearbox section, and other parts.

4. The method for intelligently generating an engine inspection report according to claim 3, characterized in that: The specific process of filling several pieces of data corresponding to the same classification field value into the same part information table template is as follows: S01. The part information table template includes a title module and an adjustable module. The title module includes a text label to be filled in, and the current classification field value replaces the text label to be filled in; S02, according to the n pieces of data corresponding to the current classification field value, the number of areas to be filled in the adjustable module in the part information table template is set to n, and each area to be filled in includes a picture label and a text label to be filled; S03, searching for the corresponding label type to be filled in the area to be filled according to the field name for each piece of data and replacing the corresponding label to be filled; S04. Adaptively create a table, layout and display the n areas to be filled in the table, and finally obtain a part information table corresponding to the current classification field value.

5. The method for intelligently generating an engine inspection report according to claim 4, characterized in that: The specific process of adaptively creating a table and typeset n areas to be filled in the table is as follows: The display area size of each area to be filled in is set to the same size, and the n areas to be filled in are arranged and displayed in a row of two at a time. The final size of the table for the layout display of the adjustable module is: 2×[n / 2], where [n / 2] is rounded up. When n is an even number, the table size is n / 2 rows and 2 columns. When n is an odd number, the table size is (n+1) / 2 rows and 2 columns.

6. The method for intelligently generating an engine inspection report according to claim 4, characterized in that: When n pieces of data are filled into the area to be filled in and displayed in a table, the n pieces of data are screened again according to the part number, and several pieces of data with the same part number are displayed in a continuous layout. When there are two pieces of data with the same part number and the corresponding fault pictures include an overall picture of the part and a partial picture of the fault, the two pieces of data containing the overall picture of the part and the partial picture of the fault are displayed side by side in the same row, and an identification symbol is added to point from the overall picture of the part to the partial picture of the fault.

7. The method for intelligently generating an engine inspection report according to claim 4, characterized in that: The text label includes a first text label and a second text label. When each piece of data is filled into each area to be filled, when the image label to be filled is replaced with the image of the fault image field, the image of the fault image field is adjusted to a uniform preset size and then filled into the image label position; When the field value of the fault description field is used to replace the position corresponding to the fault description in the first text label, the preset professional corpus is called to translate the field value of the fault description field, and the translated content replaces the position corresponding to the fault description in the second text label.

8. The method for intelligently generating an engine inspection report according to claim 4, characterized in that: The method of filling the fault information table in step S1 is: Import the fault picture into the picture field of the information table, call the part classification model to identify the parts contained in the picture, find the corresponding classification and part number in the parts library according to the part name, and automatically fill the recognition result into the part name, classification field and part number field. Call the part fault recognition model to automatically identify the fault type in the fault picture and mark the fault location in the fault picture, generate a fault description and fill it into the fault description field and refresh the fault picture field with the overall picture of the part with annotations. At the same time, extract the fault local picture from the overall picture of the part with annotations and insert it into the table below the fault picture, and copy the values ​​of other fields in the row where the overall picture of the part with annotations is located.

9. An intelligent generation system for engine inspection reports, characterized in that: A method for intelligently generating an engine inspection report according to any one of claims 1 to 8, comprising: Fault preprocessing module, used to obtain fault pictures and import them into the fault information table in batches, and fill in the fault information according to the parts information in the fault pictures; The inspection report analysis module is used to call the inspection report template and traverse each part to be replaced in the inspection report template, each part to be replaced includes a plurality of labels to be filled, and the types of labels to be filled include text or table; The inspection report generation module is used to fill the corresponding field value in the fault information table into the text label to be filled when the label type in the part to be replaced is text and the label name is the same as the field name in the fault information table; When the tag type in the part to be replaced is a table, classification is performed according to the classification field value of the field name classification in the fault information table, and each row of data in the fault information table is taken as a piece of data, and finally several pieces of data corresponding to the same classification field value are obtained; Call several parts information table templates, fill several pieces of data corresponding to the same classification field value into the same parts information table template, and finally obtain several parts information tables, and fill the several parts information tables into the table labels to be filled; The inspection report output module is used to output the engine inspection report after traversing all the parts to be replaced in the inspection report template.

10. A computer-readable storage medium, characterized in that: include: one or more processors; A storage unit, used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement an engine inspection report intelligent generation method as described in any one of claims 1-8.

Images