Picture production process management system based on digital media

Abstract

The invention discloses a picture production process management system based on digital media, which belongs to the technical field of digital media and comprises a personnel information management module, a task allocation decision module, a detection sorting module and a drawing detection module. The personnel information management module inputs multi-aspect information of personnel; the task allocation decision module screens personnel according to task types and styles, determines an allocation priority value in combination with experience and ability of the personnel, realizes accurate task allocation, improves the completion efficiency and guarantees the project progress; the detection sorting module comprehensively considers poor quality image delivery, quality image key values and quality image defective product values to determine quality image priority values, accurately sorts the images, and preferentially carries out quality inspection on key images; the drawing detection module quantitatively detects pictures from the aspects of size, definition and content integrity, objectively judges the quality, marks and feeds back unqualified pictures, promotes production personnel to improve and improve the overall quality of the pictures, and effectively solves many problems existing in traditional picture production process management.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital media, and in particular relates to a picture production process management system based on digital media. Background Art

[0002] In the era of booming digital media, image production is widely used in many fields such as e-commerce and film and television, and the market's requirements for its efficiency and quality are constantly increasing; However, the current management of image production processes has obvious shortcomings: personnel management lacks systematicity, and task allocation is often unreasonable. For example, the task of producing retro-style advertising images may be assigned to personnel who are not proficient in this style, resulting in low production efficiency and high costs. During task execution, there is a lack of effective tracking and evaluation of progress and personnel status, making it difficult to solve problems in a timely manner; image quality detection relies on subjective experience, lacks unified standards and quantitative methods, and is difficult to accurately locate problems. For this reason, we propose a digital media-based image production process management system. Summary of the Invention

[0003] The purpose of the present invention is to provide a digital media-based picture production process management system to solve the problems raised in the above background technology.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a digital media-based picture production process management system, comprising: a personnel information management module, a task allocation decision module, a detection and sorting module, and a mapping detection module; The personnel information management module is used to enter personnel information; The task allocation decision module is used to screen personnel from the personnel information database according to the production type and production style of the new picture production task when a new picture production task is issued, and then analyze the allocation priority values ​​of the screened personnel and allocate tasks according to the allocation priority values ​​of the screened personnel; The inspection and sorting module is used to collect completed production images and record them as quality inspection images, analyze the quality image priority values ​​corresponding to the quality inspection images, and sort all quality inspection images according to the quality image priority values ​​corresponding to each quality inspection image to obtain the image quality inspection sorting sequence; The mapping inspection module conducts quality inspection according to the image quality inspection sorting sequence, and checks the size, clarity, and content completeness of the quality inspection images to determine whether the quality inspection images are qualified.

[0005] Preferably, the specific process of personnel information entry by the personnel information management module is as follows: Establish a personnel information database and enter the personnel information of all image production personnel, including: name, contact information, email address, skills, task history, and task log; Skills include: proficiency in image production types and styles; The task history includes: the category, name, task scheduling time, task completion time, and number of unqualified images for all tasks completed in the past; The task log includes: the number of current mapping tasks undertaken, the estimated completion time, and the task progress.

[0006] Preferably, the specific process of the task allocation decision module to screen preliminary personnel from the personnel information database according to the production type and production style of the new picture production task is as follows: Obtain task types and image style requirements for new image creation tasks. Task types include: e-commerce, film and television, advertising, brand promotion, and illustration. Image style requirements include: Chinese style, retro style, minimalist style, European and American style, Japanese style, cyberpunk style, realistic style, and cartoon style. For each staff member, obtain the image production category that the staff member is good at from the personnel information database, and compare the image production category that the staff member is good at with the corresponding task type in the new image production task. If the image production category that the staff member is good at contains an image production category that is consistent with the new image production task type, then the staff member is marked as a preliminary screening personnel; For the initially screened personnel, the various picture production styles they are good at are compared one by one, and the personnel who have the picture style required for the new picture production task are selected, and these initially screened personnel are recorded as preliminary selected personnel.

[0007] Preferably, the task allocation decision module analyzes the allocation priority values ​​of the preliminary candidates and allocates tasks according to the allocation priority values ​​of the preliminary candidates as follows: For each primary candidate, extract all the picture making tasks that the primary candidate has completed in the past that have the same task type and picture style as the new picture making task from the personnel information database, record these picture making tasks as homogeneous tasks, and count the number of all homogeneous tasks, which is recorded as the total number of homogeneous tasks RS; For each homogeneous task, the task time RT is obtained by subtracting the task scheduling time from the task completion time. The average task time of all homogeneous tasks is calculated and recorded as the average task time RJ using the formula: , get the Renbo value RP, where RTi represents the task time of the i-th homogeneous task, i=1, 2, ..., RS; after normalizing the task average consumption RJ and the Renbo value RP, use the formula: RN=RJ×a1-RP×a2 to get the task efficiency value RN, where a1 and a2 are preset weight coefficients; Obtain the number of mapping tasks currently undertaken by the preliminary members from the task log of the preliminary members, recorded as the current task number DS, as well as the task progress and expected completion time corresponding to each currently undertaken mapping task. By accumulating the task progress and expected completion time corresponding to all currently undertaken mapping tasks and dividing them by the current task number DS, we can obtain the current average task progress XD and the current average completion time XT. After normalizing the task number DS, the current average task progress XD and the current average completion time XT, we use the formula: , get the capacity value CN, where b1, b2, and b3 are preset weight coefficients; For each candidate, after normalizing their corresponding homogeneous total number RS, task efficiency value RN, and responsibility capacity value CN, the allocation priority value FY is obtained using the formula: FY=RS×w1+RN×w2+CN×w3, where w1, w2, and w3 are preset weight coefficients; Compare the allocation priority values ​​of all the preliminary candidates, select the preliminary candidate with the largest allocation priority value FY, determine this selected preliminary candidate as the task undertaker, assign the new picture production task to the task undertaker, and indicate the delivery time of the new picture production task.

[0008] Preferably, the specific process of the detection sorting module analyzing the quality image priority value corresponding to the quality inspection image is: Establish an image quality inspection library. When staff complete an image production task, they will send the finished image to the image evaluation library for storage. When storing the image, the image production information will be marked and stored simultaneously. The image production information includes: the image name, production type, production style, delivery time, and the name, contact information, and email address of the person responsible for production. Mark the images in the image quality inspection library as quality inspection images. For each quality inspection image, obtain its corresponding mapping task delivery time. Subtract the current time from the mapping task delivery time to obtain the quality image delivery difference ZC. Establish an image type criticality mapping table. Each image production type in the mapping table corresponds to a criticality value. Match the production type corresponding to the quality inspection image with the image type criticality mapping table, and output the corresponding criticality value, which is recorded as the quality image criticality value ZG. Obtain the name of the person responsible for producing the quality inspection image, obtain the task history of the person responsible for producing from the personnel information management module based on the name of the person responsible for producing, and count the number of unqualified images produced by the person in the past from the task history, and record it as the quality image defective value ZP; After normalizing the quality image delivery difference ZC, quality image key value ZG and quality image defective value ZP corresponding to the quality inspection image, the quality image priority value ZY is obtained using the formula: ZY=ZC×z1+ZG×z2+ZP×z3, where z1, z2 and z3 are preset weight coefficients.

[0009] Preferably, the detection sorting module sorts all the quality inspection images according to the quality image priority value corresponding to each quality inspection image, and the specific process of obtaining the image quality inspection sorting sequence is as follows: For the quality inspection images in the image quality inspection library, they are sorted from high to low according to the size of the quality image priority values ​​corresponding to the quality inspection images to obtain the image quality inspection sorting sequence. Among them, if the quality inspection images correspond to the same quality image priority values, a random number is assigned to each quality inspection image with the same quality image priority value with the help of a computer random number generator, and they are re-sorted based on the size of the random number on the basis of their original corresponding sorting positions.

[0010] Preferably, the process of the mapping detection module detecting the size, clarity, and content integrity of the quality inspection image is as follows: From the image quality inspection sorting sequence generated by the detection and sorting module, obtain the images that need to be quality inspected in order; For each quality inspection image, measure the width and height values ​​of the quality inspection image, and obtain the aspect ratio by dividing the height value of the quality inspection image by the width value. The preset drawing style under each type of image production corresponds to a corresponding standard width value, height value and aspect ratio; respectively calculate the difference between the width value, height value and aspect ratio of the quality inspection image and the standard width value, height value and aspect ratio corresponding to the corresponding image production type and style, and record them as width difference KC, height difference GC and aspect ratio difference BC respectively. After normalizing the width difference KC, height difference GC and aspect ratio difference BC, use the formula: RC=KC×c1+GC×c2+BC×c3 to obtain the size difference RC, where c1, c2, and c3 are preset weight coefficients. The preset drawing style under each image type corresponds to an appearance difference deviation threshold. If the appearance difference WG of the quality inspection image is greater than or equal to the appearance difference deviation threshold corresponding to the same image type and style, the quality inspection image size is marked as unqualified; The Sobel operator is used to calculate the horizontal and vertical gradient components of each pixel in the quality inspection image, which are recorded as Gx and Gy, using the formula: , get the gradient amplitude G of the pixel point, preset the drawing style under each type of picture production and correspond to the corresponding edge gradient amplitude threshold, if the gradient amplitude G of the pixel point is greater than the preset edge gradient amplitude threshold under the corresponding type and style, then the pixel point is marked as an edge pixel point, and the gradient amplitude of all edge pixels is averaged to obtain the edge intensity mean of the image frame; for the drawing style under each type of picture production, preset several edge intensity mean intervals, each edge intensity mean interval corresponds to a clarity value, where the upper limit value and the lower limit value of the edge intensity mean interval are the same. The larger the threshold value, the larger the corresponding clarity value. The edge intensity mean of the quality inspection image is matched with all the corresponding edge intensity mean intervals of the same type and style. The clarity value corresponding to the successfully matched edge intensity mean interval is output to obtain the quality image clarity value. For the drawing style under each type of image production, a quality image clarity benchmark value is preset. The quality image clarity value corresponding to the quality inspection image is compared with the quality image clarity benchmark value corresponding to the same type and style. If the quality image clarity value of the quality inspection image is less than the quality image clarity benchmark value under the corresponding type and style, the quality inspection image clarity is marked as unqualified. The quality inspection image is evenly divided into several inspection areas. For each inspection area, each pixel point in it is traversed in turn to obtain its pixel value. Then, according to the variance calculation formula, the pixel values ​​corresponding to all pixels in the inspection area are calculated to obtain the pixel variance of the inspection area; the pixel variances of all inspection areas are added up and divided by the total number of inspection areas to obtain the quality map mean difference; the mapping style under each type of image production is preset to have a corresponding threshold coefficient, and the quality map mean difference of the quality inspection image is obtained by multiplying the threshold coefficient under the corresponding image production type and style. The variance threshold corresponding to the inspection image is used, and the inspection area corresponding to the pixel variance less than the corresponding variance threshold is recorded as a suspected incomplete area. The number of suspected incomplete areas is counted, and their proportion to the total number of inspection areas is calculated to obtain the content missing ratio. For the mapping style under each type of image production, a content missing ratio threshold is preset, and the content missing ratio corresponding to the quality inspection image is compared with the content missing ratio threshold under the corresponding type and style. If the content missing ratio of the quality inspection image is greater than or equal to the content missing ratio threshold under the corresponding type and style, the content richness of the quality inspection image is marked as unqualified.

[0011] Preferably, the specific process of the drawing detection module to determine whether the quality inspection picture is qualified is: If any of the three inspection indicators of size, clarity, and content richness of the quality inspection image fails to meet the qualified standards, the quality inspection image will be judged as unqualified. At the same time, the number of unqualified images produced by the person responsible for the image production will be updated in the corresponding personnel information record. The unqualified quality inspection image will be resent to the email address of the person responsible for the image production. When sending, the specific inspection indicators involved in the unqualified image production will be marked, and the person responsible for the image production will be required to make corrections based on the inspection indicators involved in the unqualified image production. If the quality inspection pictures meet the qualified standards in terms of size, clarity and content richness, the quality inspection pictures are judged to be qualified.

[0012] Compared with the prior art, the present invention has the following beneficial effects: (1) The task allocation decision module of the digital media-based picture production process management system comprehensively considers the type and style of new picture production tasks, as well as the relevant experience and capabilities of the personnel; by screening personnel with relevant task experience and combining the number of homogeneous tasks, task efficiency value and responsibility value to determine the allocation priority value, it can accurately allocate tasks to the most suitable personnel; reasonable task allocation enables personnel to give full play to their professional advantages and reduce the time of exploration and trial caused by unfamiliarity with task types or styles; at the same time, taking into account the responsibility value of personnel, it avoids the task allocation being too concentrated on a few people, ensuring that the task volume of each person is relatively balanced; this not only improves the completion speed of a single task, but also makes the progress of the entire project smoother and reduces the occurrence of task delays.

[0013] (2) The inspection and sorting module of the digital media-based image production process management system determines the quality image priority value by comprehensively considering factors such as quality image delivery difference, quality image key value, and quality image defective value. It can accurately identify urgent, important, or quality-prone images and place them in the priority quality inspection position; this ensures that those images that have a greater impact on project progress or quality can be inspected and processed in a timely manner, avoiding project delays or quality problems caused by untimely quality inspection; when encountering the same quality image priority value, the system uses random numbers to re-sort and ensure the randomness of quality inspection; this avoids the fixed sorting mode that may appear in some cases, so that each image has an equal chance of being inspected first, improving the fairness and accuracy of quality inspection.

[0014] (3) The digital media-based image production process management system, the drawing detection module performs quantitative detection on the image from three aspects: size, clarity, and content completeness. This makes the image quality detection no longer rely on subjective experience, but has an objective and accurate judgment basis. For unqualified images detected, the system will mark them and feedback them to the production staff for rectification. The production staff can make targeted improvements based on the problems reported by the system to avoid the same problems in subsequent production; BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0017] Example 1 See also Figure 1 , the present invention provides a picture production process management system based on digital media, including: a personnel information management module, a task allocation decision module, a detection and sorting module, and a mapping detection module; The personnel information management module is used to enter personnel information. The specific process is as follows: Establish a personnel information database and enter the personnel information of all image production personnel, including: name, contact information, email address, skills, task history, and task log; Skills include: proficiency in image production types and styles; The task history includes: the category, name, task scheduling time, task completion time, and number of unqualified images for all tasks completed in the past; The task log includes: the number of current mapping tasks undertaken, the estimated completion time, and the task progress; The task allocation decision module is used to screen the preliminary candidates from the personnel information database according to the production type and production style of the new picture production task when a new picture production task is issued. Then, the allocation priority values ​​of the preliminary candidates are analyzed and the task is allocated according to the allocation priority values ​​of the preliminary candidates. The specific process is as follows: Obtain task types and image style requirements for new image creation tasks. Task types include: e-commerce, film and television, advertising, brand promotion, illustration, etc.; image style requirements include: Chinese style, retro style, minimalist style, European and American style, Japanese style, cyberpunk style, realistic style, cartoon style, etc. For each staff member, obtain the image production category that the staff member is good at from the personnel information database, and compare the image production category that the staff member is good at with the corresponding task type in the new image production task. If the image production category that the staff member is good at contains an image production category that is consistent with the new image production task type, then the staff member is marked as a preliminary screening personnel; For the initially screened personnel, compare the various picture production styles they are good at one by one, select the personnel who have the picture style required for the new picture production task, and record these initially screened personnel as preliminary selected personnel; For each primary candidate, extract all the picture making tasks that the primary candidate has completed in the past that have the same task type and picture style as the new picture making task from the personnel information database, record these picture making tasks as homogeneous tasks, and count the number of all homogeneous tasks, which is recorded as the total number of homogeneous tasks RS; For each homogeneous task, the task time RT is obtained by subtracting the task scheduling time from the task completion time. The average task time of all homogeneous tasks is calculated and recorded as the average task time RJ using the formula: , we get the Renbo value RP, where RTi represents the task time of the i-th homogeneous task, i=1, 2, …, RS; after normalizing the task average time RJ and the Renbo value RP, we use the formula: RN=RJ×a1-RP×a2 to get the task efficiency value RN, where a1 and a2 are preset weight coefficients. The larger the task efficiency value, the better the comprehensive performance of the preliminary candidates in terms of time control and stability when completing tasks of the same type and style as the new task; Obtain the number of mapping tasks currently undertaken by the preliminary members from the task log of the preliminary members, recorded as the current task number DS, as well as the task progress and expected completion time corresponding to each currently undertaken mapping task. By accumulating the task progress and expected completion time corresponding to all currently undertaken mapping tasks and dividing them by the current task number DS, we can obtain the current average task progress XD and the current average completion time XT. After normalizing the task number DS, the current average task progress XD and the current average completion time XT, we use the formula: , and obtain the capacity value CN, where b1, b2, and b3 are preset weight coefficients. The larger the capacity value corresponding to the primary selected personnel, the stronger their ability to undertake new mapping tasks; For each candidate, after normalizing their corresponding homogeneous total number RS, task efficiency value RN, and responsibility capacity value CN, the allocation priority value FY is obtained using the formula: FY=RS×w1+RN×w2+CN×w3, where w1, w2, and w3 are preset weight coefficients; Compare the allocation priority values ​​of all the preliminary candidates, select the preliminary candidate with the largest allocation priority value FY, determine this selected preliminary candidate as the task undertaker, assign the new picture production task to the task undertaker, and indicate the delivery time of the new picture production task; It should be noted that by selecting preliminary screening personnel from the personnel information database based on the task type, and then determining the preliminary candidates based on the picture style, we ensure that the personnel participating in the task are highly compatible with the new task in terms of professional fields and style preferences; then we use the total number of homogeneous personnel, task efficiency value and ability to undertake value to comprehensively and quantitatively evaluate the preliminary candidates, among which the total number of homogeneous personnel reflects the level of experience of personnel in similar tasks; the task efficiency value comprehensively considers the average and fluctuation values ​​of task time, reflecting the performance of personnel in time control and stability; the ability to undertake value combines the current number of tasks undertaken, task progress and expected completion time to evaluate the ability of personnel to undertake new tasks; the total number of homogeneous personnel, task efficiency value and ability to undertake value corresponding to the preliminary candidates are comprehensively considered to obtain the corresponding allocation priority value of the preliminary candidates, and selecting the preliminary candidates with the largest allocation priority value for task allocation can improve task completion efficiency, reduce delays and rework, and make project progress more controllable.

[0018] The inspection and sorting module is used to collect completed production images and record them as quality inspection images, analyze the quality image priority values ​​corresponding to the quality inspection images, and sort all quality inspection images according to the quality image priority values ​​corresponding to each quality inspection image to obtain the image quality inspection sorting sequence. The specific process is as follows: Establish an image quality inspection library. When staff complete an image production task, they will send the finished image to the image evaluation library for storage. When storing the image, the image production information will be marked and stored simultaneously. The image production information includes: the image name, production type, production style, delivery time, and the name, contact information, and email address of the person responsible for production. Mark the images in the image quality inspection library as quality inspection images. For each quality inspection image, obtain its corresponding mapping task delivery time. Subtract the current time from the mapping task delivery time to obtain the quality image delivery difference ZC. The smaller the delivery difference, the more urgent the mapping task, and the higher the quality inspection priority of the image. Establish an image type criticality mapping table. Each image production type in the mapping table corresponds to a criticality value. Match the production type corresponding to the quality inspection image with the image type criticality mapping table, and output the corresponding criticality value, which is recorded as the quality image key value ZG. The higher the quality image key value corresponding to the quality inspection image, the more important the image production task is, and the higher the quality inspection priority of the image. Obtain the name of the person responsible for producing the quality inspection image. Based on the name, obtain the task history of the person from the personnel information management module. Count the number of unqualified images produced by the person from the task history and record it as the quality image defective value ZP. The larger the quality image defective value corresponding to the quality inspection image, the more likely the image will have problems in quality inspection, and the higher the priority for quality inspection; After normalizing the quality image delivery difference ZC, quality image key value ZG, and quality image defective value ZP corresponding to the quality inspection image, the quality image priority value ZY is obtained using the formula: ZY=ZC×z1+ZG×z2+ZP×z3, where z1, z2, and z3 are preset weight coefficients; For the quality inspection images in the image quality inspection library, sort them from high to low according to the size of their corresponding quality image priority values ​​to obtain an image quality inspection sorting sequence. If the quality inspection images have the same quality image priority values, a random number is assigned to each of them using a computer random number generator, and they are re-sorted based on their original corresponding sorting positions according to the size of the random number. It should be noted that, during sorting, key parameters such as quality image delivery difference, quality image key value, and quality image defective value are integrated to accurately measure the urgency, importance, and potential quality risk of each quality inspection image, so as to achieve accurate and efficient sorting. When encountering the situation where the quality image priority values ​​are the same, the computer random number generator is used to re-sort them based on the original sorting position, which not only ensures randomness but also takes into account continuity, effectively resolves the sorting problems under special circumstances, and allows the sorting rules to be flexibly applied in complex and changeable actual scenarios; at the same time, through the normalized comprehensive analysis of quality image delivery difference, quality image key value, and quality image defective value, the quality image priority value is obtained, which gets rid of the interference of subjective judgment, provides a scientific quantitative basis for sorting, and ensures that the sorting is fair and objective.

[0019] The drawing inspection module conducts quality inspection according to the image quality inspection sorting sequence, and checks the size, clarity, and content completeness of the quality inspection images to determine whether the quality inspection images are qualified. The specific process is as follows: From the image quality inspection sorting sequence generated by the detection and sorting module, obtain the images that need to be quality inspected in order; For each quality inspection image, measure the width and height values ​​of the quality inspection image, and obtain the aspect ratio by dividing the height value of the quality inspection image by the width value. The preset drawing style under each type of image production corresponds to a corresponding standard width value, height value and aspect ratio; respectively calculate the difference between the width value, height value and aspect ratio of the quality inspection image and the standard width value, height value and aspect ratio corresponding to the corresponding image production type and style, and record them as width difference KC, height difference GC and aspect ratio difference BC respectively. After normalizing the width difference KC, height difference GC and aspect ratio difference BC, use the formula: RC=KC×c1+GC×c2+BC×c3 to obtain the size difference RC, where c1, c2, and c3 are preset weight coefficients. The preset drawing style under each image type corresponds to an appearance difference deviation threshold. If the appearance difference WG of the quality inspection image is greater than or equal to the appearance difference deviation threshold corresponding to the same image type and style, the quality inspection image size is marked as unqualified; The Sobel operator is used to calculate the horizontal and vertical gradient components of each pixel in the quality inspection image, which are recorded as Gx and Gy, using the formula: , get the gradient amplitude G of the pixel point, preset the drawing style under each type of picture production and correspond to the corresponding edge gradient amplitude threshold, if the gradient amplitude G of the pixel point is greater than the preset edge gradient amplitude threshold under the corresponding type and style, then the pixel point is marked as an edge pixel point, and the gradient amplitude of all edge pixels is averaged to obtain the edge intensity mean of the image frame; for the drawing style under each type of picture production, preset several edge intensity mean intervals, each edge intensity mean interval corresponds to a clarity value, where the upper limit value and the lower limit value of the edge intensity mean interval are the same. The larger the threshold value, the larger the corresponding clarity value. The edge intensity mean of the quality inspection image is matched with all the corresponding edge intensity mean intervals of the same type and style. The clarity value corresponding to the successfully matched edge intensity mean interval is output to obtain the quality image clarity value. For the drawing style under each type of image production, a quality image clarity benchmark value is preset. The quality image clarity value corresponding to the quality inspection image is compared with the quality image clarity benchmark value corresponding to the same type and style. If the quality image clarity value of the quality inspection image is less than the quality image clarity benchmark value under the corresponding type and style, the quality inspection image clarity is marked as unqualified. The quality inspection image is evenly divided into several inspection areas. For each inspection area, each pixel point in it is traversed in turn to obtain its pixel value. Then, according to the variance calculation formula, the pixel values ​​corresponding to all pixels in the inspection area are calculated to obtain the pixel variance of the inspection area; the pixel variances of all inspection areas are added up and divided by the total number of inspection areas to obtain the quality map mean difference; the mapping style under each type of image production is preset to have a corresponding threshold coefficient, and the quality map mean difference of the quality inspection image is obtained by multiplying the threshold coefficient under the corresponding image production type and style. The variance threshold corresponding to the inspection image is used. The inspection area corresponding to the pixel variance less than the corresponding variance threshold is recorded as a suspected incomplete area. The number of suspected incomplete areas is counted and their proportion to the total number of inspection areas is calculated to obtain the content missing ratio. For each mapping style under the type of image production, a content missing ratio threshold is preset. The content missing ratio corresponding to the inspection image is compared with the content missing ratio threshold under the corresponding type and style. If the content missing ratio of the inspection image is greater than or equal to the content missing ratio threshold under the corresponding type and style, the inspection image is marked as unqualified for content richness. If any of the three inspection indicators of size, clarity, and content richness of the quality inspection image fails to meet the qualified standards, the quality inspection image will be judged as unqualified. At the same time, the number of unqualified images produced by the person responsible for the image production will be updated in the corresponding personnel information record. The unqualified quality inspection image will be resent to the email address of the person responsible for the image production. When sending, the specific inspection indicators involved in the unqualified image production will be marked, and the person responsible for the image production will be required to make corrections based on the inspection indicators involved in the unqualified image production. If the quality inspection pictures meet the qualified standards in terms of size, clarity and content richness, the quality inspection pictures are judged to be qualified.

[0020] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A digital media-based image production process management system, including: Personnel information management module, task allocation decision module, detection and sorting module, and mapping and detection module; characterized by: The personnel information management module is used to enter personnel information; The task allocation decision module is used to screen personnel from the personnel information database according to the production type and production style of the new picture production task when a new picture production task is issued, and then analyze the allocation priority values ​​of the screened personnel and allocate tasks according to the allocation priority values ​​of the screened personnel; The inspection and sorting module is used to collect completed production images and record them as quality inspection images, analyze the quality image priority values ​​corresponding to the quality inspection images, and sort all quality inspection images according to the quality image priority values ​​corresponding to each quality inspection image to obtain the image quality inspection sorting sequence; The mapping inspection module conducts quality inspection according to the image quality inspection sorting sequence, and checks the size, clarity, and content completeness of the quality inspection images to determine whether the quality inspection images are qualified.

2. The digital media-based picture production process management system according to claim 1, characterized in that: The specific process of personnel information entry in the personnel information management module is as follows: Establish a personnel information database and enter the personnel information of all image production personnel, including: name, contact information, email address, skills, task history, and task log; Skills include: proficiency in image production types and styles; The task history includes: the category, name, task scheduling time, task completion time, and number of unqualified images for all tasks completed in the past; The task log includes: the number of current mapping tasks undertaken, the estimated completion time, and the task progress.

3. The digital media-based picture production process management system according to claim 2, characterized in that: The specific process of the task allocation decision module to screen the initial candidates from the personnel information database according to the production type and production style of the new image production task is as follows: Obtain task types and image style requirements for new image creation tasks. Task types include: e-commerce, film and television, advertising, brand promotion, and illustration. Image style requirements include: Chinese style, retro style, minimalist style, European and American style, Japanese style, cyberpunk style, realistic style, and cartoon style. For each staff member, obtain the image production category that the staff member is good at from the personnel information database, and compare the image production category that the staff member is good at with the corresponding task type in the new image production task. If the image production category that the staff member is good at contains an image production category that is consistent with the new image production task type, then the staff member is marked as a preliminary screening personnel; For the initially screened personnel, the various picture production styles they are good at are compared one by one, and the personnel who have the picture style required for the new picture production task are selected, and these initially screened personnel are recorded as preliminary selected personnel.

4. The digital media-based picture production process management system according to claim 3, characterized in that: The task allocation decision module analyzes the allocation priority values ​​of the preliminary candidates and allocates tasks based on the allocation priority values ​​of the preliminary candidates. The specific process is as follows: For each primary candidate, extract all the picture making tasks that the primary candidate has completed in the past that have the same task type and picture style as the new picture making task from the personnel information database, record these picture making tasks as homogeneous tasks, and count the number of all homogeneous tasks, which is recorded as the total number of homogeneous tasks RS; For each homogeneous task, the task time RT is obtained by subtracting the task scheduling time from the task completion time. The average task time of all homogeneous tasks is calculated and recorded as the average task time RJ using the formula: , get the Renbo value RP, where RTi represents the task time of the i-th homogeneous task, i=1, 2, ..., RS; after normalizing the task average consumption RJ and the Renbo value RP, use the formula: RN=RJ×a1-RP×a2 to get the task efficiency value RN, where a1 and a2 are preset weight coefficients; Obtain the number of mapping tasks currently undertaken by the preliminary members from the task log of the preliminary members, recorded as the current task number DS, as well as the task progress and expected completion time corresponding to each currently undertaken mapping task. By accumulating the task progress and expected completion time corresponding to all currently undertaken mapping tasks and dividing them by the current task number DS, we can obtain the current average task progress XD and the current average completion time XT. After normalizing the task number DS, the current average task progress XD and the current average completion time XT, we use the formula: , get the capacity value CN, where b1, b2, and b3 are preset weight coefficients; For each candidate, after normalizing their corresponding homogeneous total number RS, task efficiency value RN, and responsibility capacity value CN, the allocation priority value FY is obtained using the formula: FY=RS×w1+RN×w2+CN×w3, where w1, w2, and w3 are preset weight coefficients; Compare the allocation priority values ​​of all the preliminary candidates, select the preliminary candidate with the largest allocation priority value FY, determine this selected preliminary candidate as the task undertaker, assign the new picture production task to the task undertaker, and indicate the delivery time of the new picture production task.

5. The digital media-based picture production process management system according to claim 4, characterized in that: The specific process of the detection and sorting module analyzing the quality image priority value corresponding to the quality inspection image is as follows: Establish an image quality inspection library. When staff complete an image production task, they will send the finished image to the image evaluation library for storage. When storing the image, the image production information will be marked and stored simultaneously. The image production information includes: the image name, production type, production style, delivery time, and the name, contact information, and email address of the person responsible for production. Mark the images in the image quality inspection library as quality inspection images. For each quality inspection image, obtain its corresponding mapping task delivery time. Subtract the current time from the mapping task delivery time to obtain the quality image delivery difference ZC. Establish an image type criticality mapping table. Each image production type in the mapping table corresponds to a criticality value. Match the production type corresponding to the quality inspection image with the image type criticality mapping table, and output the corresponding criticality value, which is recorded as the quality image criticality value ZG. Obtain the name of the person responsible for producing the quality inspection image, obtain the task history of the person responsible for producing from the personnel information management module based on the name of the person responsible for producing, and count the number of unqualified images produced by the person in the past from the task history, and record it as the quality image defective value ZP; After normalizing the quality image delivery difference ZC, quality image key value ZG and quality image defective value ZP corresponding to the quality inspection image, the quality image priority value ZY is obtained using the formula: ZY=ZC×z1+ZG×z2+ZP×z3, where z1, z2 and z3 are preset weight coefficients.

6. The digital media-based picture production process management system according to claim 5, characterized in that: The detection sorting module sorts all the quality inspection images according to the quality image priority value corresponding to each quality inspection image. The specific process of obtaining the image quality inspection sorting sequence is as follows: For the quality inspection images in the image quality inspection library, they are sorted from high to low according to the size of the quality image priority values ​​corresponding to the quality inspection images to obtain the image quality inspection sorting sequence. Among them, if the quality inspection images correspond to the same quality image priority values, a random number is assigned to each quality inspection image with the same quality image priority value with the help of a computer random number generator, and they are re-sorted based on the size of the random number on the basis of their original corresponding sorting positions.

7. The digital media-based picture production process management system according to claim 6, characterized in that: The process of the mapping detection module to detect the size, clarity, and content integrity of the quality inspection image is as follows: From the image quality inspection sorting sequence generated by the detection and sorting module, obtain the images that need to be quality inspected in order; For each quality inspection image, measure the width and height values ​​of the quality inspection image, and obtain the aspect ratio by dividing the height value of the quality inspection image by the width value. The preset drawing style under each type of image production corresponds to a corresponding standard width value, height value and aspect ratio; respectively calculate the difference between the width value, height value and aspect ratio of the quality inspection image and the standard width value, height value and aspect ratio corresponding to the corresponding image production type and style, and record them as width difference KC, height difference GC and aspect ratio difference BC respectively. After normalizing the width difference KC, height difference GC and aspect ratio difference BC, use the formula: RC=KC×c1+GC×c2+BC×c3 to obtain the size difference RC, where c1, c2, and c3 are preset weight coefficients. The preset drawing style under each image type corresponds to an appearance difference deviation threshold. If the appearance difference WG of the quality inspection image is greater than or equal to the appearance difference deviation threshold corresponding to the same image type and style, the quality inspection image size is marked as unqualified; The Sobel operator is used to calculate the horizontal and vertical gradient components of each pixel in the quality inspection image, which are recorded as Gx and Gy, using the formula: , get the gradient amplitude G of the pixel point, preset the drawing style under each type of picture production and correspond to the corresponding edge gradient amplitude threshold, if the gradient amplitude G of the pixel point is greater than the preset edge gradient amplitude threshold under the corresponding type and style, then the pixel point is marked as an edge pixel point, and the gradient amplitude of all edge pixels is averaged to obtain the edge intensity mean of the image frame; for the drawing style under each type of picture production, preset several edge intensity mean intervals, each edge intensity mean interval corresponds to a clarity value, where the upper limit value and the lower limit value of the edge intensity mean interval are the same. The larger the threshold value, the larger the corresponding clarity value. The edge intensity mean of the quality inspection image is matched with all the corresponding edge intensity mean intervals of the same type and style. The clarity value corresponding to the successfully matched edge intensity mean interval is output to obtain the quality image clarity value. For the drawing style under each type of image production, a quality image clarity benchmark value is preset. The quality image clarity value corresponding to the quality inspection image is compared with the quality image clarity benchmark value corresponding to the same type and style. If the quality image clarity value of the quality inspection image is less than the quality image clarity benchmark value under the corresponding type and style, the quality inspection image clarity is marked as unqualified. The quality inspection image is evenly divided into several inspection areas. For each inspection area, each pixel point in it is traversed in turn to obtain its pixel value. Then, according to the variance calculation formula, the pixel values ​​corresponding to all pixels in the inspection area are calculated to obtain the pixel variance of the inspection area; the pixel variances of all inspection areas are added up and divided by the total number of inspection areas to obtain the quality map mean difference; the mapping style under each type of image production is preset to have a corresponding threshold coefficient, and the quality map mean difference of the quality inspection image is obtained by multiplying the threshold coefficient under the corresponding image production type and style. The variance threshold corresponding to the inspection image is used, and the inspection area corresponding to the pixel variance less than the corresponding variance threshold is recorded as a suspected incomplete area. The number of suspected incomplete areas is counted, and their proportion to the total number of inspection areas is calculated to obtain the content missing ratio. For the mapping style under each type of image production, a content missing ratio threshold is preset, and the content missing ratio corresponding to the quality inspection image is compared with the content missing ratio threshold under the corresponding type and style. If the content missing ratio of the quality inspection image is greater than or equal to the content missing ratio threshold under the corresponding type and style, the content richness of the quality inspection image is marked as unqualified.

8. The digital media-based picture production process management system according to claim 7, characterized in that: The specific process of the drawing detection module to determine whether the quality inspection picture production is qualified is as follows: If any of the three inspection indicators of size, clarity, and content richness of the quality inspection image fails to meet the qualified standards, the quality inspection image will be judged as unqualified. At the same time, the number of unqualified images produced by the person responsible for the image production will be updated in the corresponding personnel information record. The unqualified quality inspection image will be resent to the email address of the person responsible for the image production. When sending, the specific inspection indicators involved in the unqualified image production will be marked, and the person responsible for the image production will be required to make corrections based on the inspection indicators involved in the unqualified image production. If the quality inspection pictures meet the qualified standards in terms of size, clarity and content richness, the quality inspection pictures are judged to be qualified.

Images