CN113643260A - Method, apparatus, device, medium and product for detecting image quality - Google Patents
Method, apparatus, device, medium and product for detecting image quality Download PDFInfo
- Publication number
- CN113643260A CN113643260A CN202110930170.XA CN202110930170A CN113643260A CN 113643260 A CN113643260 A CN 113643260A CN 202110930170 A CN202110930170 A CN 202110930170A CN 113643260 A CN113643260 A CN 113643260A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- field
- target image
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Abstract
The present disclosure provides a method, an apparatus, a device, a medium, and a product for detecting image quality, which relate to the technical field of artificial intelligence, specifically to the technical field of computer vision and deep learning, and can be applied in the scenes of image processing, image recognition, and the like. The specific implementation scheme is as follows: acquiring a target image; determining at least one model corresponding to the target image; determining image parameters corresponding to the target image based on the target image and the at least one model; the image parameters include at least one of: a definition parameter, an integrity parameter, a type parameter; based on the image parameters, an image quality result of the target image is determined. This implementation can improve image quality detection efficiency.
Description
Technical Field
The disclosure relates to the technical field of artificial intelligence, in particular to the technical field of computer vision and deep learning, and can be applied to the scenes of graphic processing, image recognition and the like.
Background
Currently, during the process of personal related information entry, a large number of text proof type images need to be collected, such as license licenses, revenue proofs, and the like. By performing character recognition on these images, personal related information is obtained.
However, the images have different quality, and the images are often incomplete and unsatisfactory, so that subsequent character recognition is affected, and the character recognition effect is poor. In contrast, a manual verification mode is usually adopted at present, and after the image is manually verified, an image with higher quality is obtained, and subsequent character recognition is performed, so that the character recognition effect is improved. However, the manual verification of images has a problem of low efficiency.
Disclosure of Invention
The present disclosure provides a method, apparatus, device, medium, and article for detecting image quality.
According to an aspect of the present disclosure, there is provided a method for detecting image quality, including: acquiring a target image; determining at least one model corresponding to the target image; determining image parameters corresponding to the target image based on the target image and the at least one model; the image parameters include at least one of: a definition parameter, an integrity parameter, a type parameter; determining an image quality result of the target image based on the image parameters.
According to another aspect of the present disclosure, there is provided an apparatus for detecting image quality, including: an image acquisition unit configured to acquire a target image; a model determination unit configured to determine at least one model corresponding to the target image; a parameter determination unit configured to determine an image parameter corresponding to the target image based on the target image and the at least one model; the image parameters include at least one of: a definition parameter, an integrity parameter, a type parameter; a quality evaluation unit configured to determine an image quality result of the target image based on the image parameter.
According to another aspect of the present disclosure, there is provided an electronic device including: one or more processors; a memory for storing one or more programs; when executed by one or more processors, cause the one or more processors to implement a method for detecting image quality as any one of the above.
According to another aspect of the present disclosure, there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method for detecting image quality as any one of the above.
According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements a method for detecting image quality as any one of the above.
According to the technology of the present disclosure, a method for detecting image quality is provided, which can improve image quality detection efficiency.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present disclosure may be applied;
FIG. 2 is a flow diagram of one embodiment of a method for detecting image quality according to the present disclosure;
FIG. 3 is a schematic diagram of one application scenario of a method for detecting image quality according to the present disclosure;
FIG. 4 is a flow diagram of another embodiment of a method for detecting image quality according to the present disclosure;
FIG. 5 is a schematic block diagram of one embodiment of an apparatus for detecting image quality according to the present disclosure;
fig. 6 is a block diagram of an electronic device for implementing a method for detecting image quality according to an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, and 103 may be electronic devices such as a mobile phone, a computer, and a tablet, and the terminal devices 101, 102, and 103 may be installed with software for character recognition, which may perform character recognition on an input image. In this embodiment, the software installed in the terminal devices 101, 102, and 103 may send the image to the server 105 through the network 104 before performing character recognition, so that the server 105 returns the image quality result. If the image quality result indicates that the image quality is better, further character recognition is carried out. If the image quality result indicates that the image quality is poor, the image is not subjected to character recognition. The process can avoid identifying images with poor quality, and improves the accuracy of character identification results.
The terminal apparatuses 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices including, but not limited to, an in-vehicle computer, an in-vehicle tablet, a vehicle control device, and the like. When the terminal apparatuses 101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.
The server 105 may be a server that provides various services, for example, the server 105 may acquire a target image transmitted by the terminal devices 101, 102, 103 through the network 104, determine at least one model corresponding to the target image, determine image parameters corresponding to the target image based on the target image and the at least one model, determine an image quality result of the target image based on the image parameters, and return the image quality result to the terminal devices 101, 102, 103.
The server 105 may be hardware or software. When the server 105 is hardware, it may be implemented as a distributed server cluster composed of a plurality of servers, or may be implemented as a single server. When the server 105 is software, it may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein.
It should be noted that the method for detecting image quality provided by the embodiment of the present disclosure may be executed by the terminal devices 101, 102, and 103, or may be executed by the server 105, and the apparatus for detecting image quality may be disposed in the terminal devices 101, 102, and 103, or may be disposed in the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to FIG. 2, a flow 200 of one embodiment of a method for detecting image quality in accordance with the present disclosure is shown. The method for detecting the image quality of the embodiment comprises the following steps:
In this embodiment, the executing entity (e.g., the terminal device 101, 102, 103 or the server 105 in fig. 1) may acquire the target image from other electronic devices that are locally stored or have a connection established in advance. The target image may be a picture or a video frame determined from various videos. For the case that the target image is a video frame determined from the video, the execution subject may extract several video frames from the video as the target image, determine an image quality result based on each extracted video frame, and finally obtain a quality evaluation result corresponding to the video in a summary manner. The number of the target images may be one, or two or more, which is not limited in this embodiment.
At step 202, at least one model corresponding to the target image is determined.
In this embodiment, at least one model is used to determine the image information of the target image from different image detection dimensions, and may include an image sharpness detection dimension, an image integrity detection dimension, an image type detection dimension, and the like. Optionally, after the execution main body obtains the target image, the evaluation requirement information corresponding to the target image may be determined first, and then at least one model corresponding to the target image is selected from the preset multiple models based on the evaluation requirement information.
The evaluation requirement information may be configured manually by a technician, or the executive subject may determine an initial classification corresponding to the image based on the image recognition, and then determine the evaluation requirement information of the image based on the initial classification. For example, the executing subject may perform image recognition on the target image, and if the initial classification of the image is recognized as the diagnosis description type, further determine at least one model corresponding to the diagnosis description type, such as a field definition evaluation model; if the initial classification of the image is identified as a certification seal type, further determining at least one model corresponding to the certification seal type, such as an integrity evaluation model; if the initial classification of the image is identified as a batch processing type, at least one model corresponding to the batch processing type, such as a whole image sharpness evaluation model, is further determined.
In this embodiment, the definition parameter is used to describe the definition of the target image, the integrity parameter is used to describe the integrity of the target image, and the type parameter is used to describe the image attribute classification of the target image, which may include but is not limited to normal shot image, copy, scan, screen flipper, etc. The initial classification is a category corresponding to the content of the target image, and the type parameter is a category corresponding to the image attribute of the target image, and the classification criteria are different. After the execution main body determines to obtain at least one model corresponding to the target image, because the at least one model is used for determining the image information of the target image from different image detection dimensions, the target image can be respectively input into the at least one model to obtain the output result of each model, and then the image parameters corresponding to different image detection dimensions are obtained based on the output result of each model. Optionally, the definition parameter may correspond to a corresponding definition evaluation model, the integrity parameter may correspond to a corresponding integrity evaluation model, and the type parameter may correspond to a corresponding type recognition model. Or, the definition parameter, the integrity parameter, and the type parameter may also correspond to the same model, and the specific corresponding relationship is not limited in this embodiment. The integrity evaluation model may be a classification model or a keypoint model, which is not limited in this embodiment.
And step 204, determining an image quality result of the target image based on the image parameters.
In this embodiment, after obtaining the image parameters, the executing subject may combine the parameters to obtain the image quality result of the target image. The image quality result is used for evaluating the image quality from the different image detection dimensions, such as image definition evaluation, image integrity evaluation and image type evaluation. The image quality result may be an evaluation result corresponding to a single target image, or may be an evaluation result corresponding to two or more target images, which is not limited in this embodiment.
In some optional implementations of the embodiment, determining the image quality result of the target image based on the image parameter may include: determining a definition grade corresponding to the target image based on the definition parameters and preset definition classification; optionally, determining an integrity level corresponding to the target image based on the integrity parameter and a preset integrity classification; optionally, determining an attribute type corresponding to the target image based on the type parameter; optionally, determining an effective field level corresponding to the target image based on the field number of the target image and the area of the field detection box, where the image parameter may further include the field number of the target image and the area of the field detection box; and generating an image quality result based on any combination of the definition level, the integrity level, the attribute type and the valid field level.
With this alternative implementation, the execution agent may preset a definition classification, such as clear and unclear, or low definition, medium definition, high definition, and an integrity classification, such as complete and incomplete, or low integrity, medium integrity, and high integrity. After the sharpness parameter is obtained, its corresponding sharpness level, e.g., sharpness high, may be determined based on the sharpness parameter. And after obtaining the integrity parameter, may determine its corresponding integrity level, such as high integrity, based on the integrity parameter. In addition, the execution subject may determine the attribute type corresponding to the target image based on analyzing the type parameter. And determining a valid field level based on the number of fields of the target image and the area of the field detection box. Specifically, the field number of the target image may be multiplied by a preset number weight, the product of the field number and the product of the area of the field detection frame multiplied by a preset area weight is added to obtain an effective field parameter, and then an effective field level corresponding to the effective field parameter is determined based on a preset level classification. Wherein, the more the number of fields, the higher the valid field parameter, and the higher the valid field level. The larger the area of the field detection box is, the larger the valid field parameter is, and the higher the valid field level is. The resulting valid field level is used to describe the presence of recognizable fields in the target image. The execution body may then generate a comprehensive image quality result based on these sharpness levels, completeness levels, attribute types, and valid field levels.
With continued reference to fig. 3, a schematic diagram of one application scenario of the method for detecting image quality according to the present disclosure is shown. In the application scenario of fig. 3, the executing subject may first obtain the target image 301, and then determine the image parameters 302 of the target image 301 based on at least one model corresponding to the target image 301, where the image parameters 302 may include, but are not limited to, a full-image sharpness score 3021, a field sharpness score 3022, an image type 3023, and an image integrity parameter 3024. The execution subject may then perform an integration process on these image parameters 302, generating and outputting an image quality result 303.
According to the method for detecting image quality provided by the above embodiment of the present disclosure, by determining at least one model corresponding to a target image, based on the target image and the at least one model, image parameters of different image detection dimensions corresponding to the target image are determined, and based on the image parameters of the different image detection dimensions, an image quality result is generated. The process realizes automatic control of image quality and can improve the image quality detection efficiency. Moreover, the image parameters of various different image detection dimensions are considered when the image quality result is determined, and the comprehensiveness and the accuracy of image quality detection can be improved.
With continued reference to fig. 4, a flow 400 of another embodiment of a method for detecting image quality according to the present disclosure is shown. As shown in fig. 4, the method for detecting image quality of the present embodiment may include the steps of:
In this embodiment, please refer to the detailed description of step 201 for the detailed description of step 401, which is not repeated herein.
Optionally, after the target image is obtained, the execution main body may further perform image preprocessing operation on the target image, and then determine an image parameter according to the image after the preprocessing operation, so as to further improve the accuracy of determining the image parameter.
At step 402, at least one model corresponding to a target image is determined.
In this embodiment, the at least one model includes a whole graph definition evaluation model, a field definition evaluation model, and an integrity evaluation model; the definition parameters at least comprise a whole image definition score and a field definition score. The integral image definition evaluation model is used for evaluating the integral image definition corresponding to the target image, the field definition evaluation model is used for evaluating the field definition corresponding to the field in the target image, and the integrity evaluation model is used for evaluating the integrity degree of the target image. The higher the overall sharpness score is, the higher the overall sharpness of the image of the target image is. The higher the field definition score, the higher the field definition indicating the target image.
For the detailed description of step 402, please refer to the detailed description of step 202, which is not repeated herein.
And step 403, determining the confidence coefficient that the target image belongs to the definition category based on the target image and the definition evaluation model of the whole image.
In this embodiment, the execution subject may input the target image into the whole-image sharpness evaluation model, and obtain confidence levels that the target image output by the whole-image sharpness evaluation model belongs to each preset sharpness classification. For example, when the entire image sharpness evaluation model is a binary classification model based on MobileNet (a convolutional neural network model), each preset sharpness classification may be sharpness and blur. At this time, the output result of the whole image definition evaluation model is the confidence degree that the target image belongs to the definition category and the confidence degree that the target image belongs to the fuzzy category.
It should be noted that the whole graph definition evaluation model may also be a multi-classification model or other forms of two-classification models, which is not limited in this embodiment.
Based on the confidence, the overall graph sharpness score is determined, step 404.
In this embodiment, the execution subject may determine the confidence that the target image belongs to the sharpness category as the overall image sharpness score. Optionally, when the whole image sharpness evaluation model is a multi-classification model, the execution subject may determine the confidence that the target image belongs to the clearest class as the whole image sharpness score, or the execution subject may also determine the whole image sharpness score in another calculation manner, which is not limited in this embodiment.
In this embodiment, the executing entity may perform image recognition on the target image by using an existing image recognition technology to obtain each target field in the recognized target image, where the target fields form the target field set.
In some optional implementations of this embodiment, determining the target field set corresponding to the target image may include: determining each candidate field corresponding to the target image and the confidence coefficient of each candidate field based on the target image and a preset character detection model; and determining a target field from each candidate field based on the confidence of each candidate field to obtain a target field set.
In this implementation, the preset character detection model is used to screen text fields in the target image. Alternatively, the text detection model may be a DB-net (differentiable binary network) based model. The execution subject may input the target image into the preset character detection model, and obtain all fields existing in the target image output by the preset character detection model, that is, the above-mentioned candidate fields, and the confidence of each candidate field. Here, the confidence of each candidate field is used to describe the probability that the identified region corresponding to the candidate field exists in the field. Then, the execution subject may select a preset number of target fields from each candidate field based on the order of the confidence degrees of each candidate field from high to low, to obtain a target field set.
And 406, determining definition scores corresponding to all the target fields in the target field set based on the target field set and the field definition evaluation model.
In this embodiment, after the execution main body obtains the target field set, since the target field set is substantially a set formed by field region images corresponding to each target field in the target image, the execution main body may input each field region image into the field definition evaluation model, and determine the definition score corresponding to each field region image, that is, the definition score corresponding to each target field. The field definition evaluation model and the entire image definition evaluation model may use the same definition evaluation model or may use definition evaluation models with different accuracies, which is not limited in this embodiment. Optionally, the field definition evaluation model may also adopt a binary model based on MobileNet, or other forms of binary models, or multi-classification models, and the like, which is not limited in this embodiment. Moreover, for the determination of the definition score corresponding to each target field, similar to the determination of the definition score of the whole image, the confidence that the field region image corresponding to each target field belongs to the definition category is determined as the definition score, which is not described herein again.
In this embodiment, after the execution main body obtains the definition scores corresponding to the target fields, the definition scores may be integrated to obtain the field definition scores. The integration process may include, but is not limited to, averaging, weighted summation, and the like, which is not limited in this embodiment.
In some optional implementation manners of this embodiment, determining the field definition score based on the definition scores corresponding to the target fields includes: and determining field definition scores based on the definition scores corresponding to the target fields and the confidence degrees of the target fields.
In this implementation, the execution subject may obtain a confidence corresponding to each target field, where the confidence corresponding to each target field is used to describe a probability that a field exists in a field region corresponding to the target field. Then, the execution body may multiply the definition score corresponding to each target field by the confidence of the target field to obtain a product corresponding to the target field, and then sum the products of the target fields to obtain a field definition score.
And step 408, determining each key point corresponding to the target image based on the target image and the integrity evaluation model.
In this embodiment, the execution subject may further input the target image into the integrity evaluation model to obtain each key point output by the integrity evaluation model after performing key point identification on the target image. The integrity evaluation model is a key point model and can be realized by adopting various existing key point models, such as a key four-corner point model, a key four-side line segment model, a key part model and the like.
In this embodiment, the execution subject may determine the integrity parameter of the target image based on a comparison result between each key point corresponding to the detected target image and each key point of a preset complete image. For example, the execution subject may determine the integrity parameter based on detecting a difference between key points corresponding to four corners of the target image and key points corresponding to four corners of a preset complete image. Wherein, the larger the difference is, the smaller the completeness parameter is, and the probability of representing that the target image is incomplete is higher.
In this embodiment, please refer to the detailed description of step 204 for the detailed description of step 410, which is not repeated herein.
In this embodiment, the execution subject may determine that the target image is an image that requires character recognition or an image that does not require character recognition based on the image quality result. The determination of the character recognition type may be performed by setting a determination condition in advance, and the character recognition type may be determined by comparing the image quality result with the determination result. For example, if the preset determination condition is that the image clarity level is higher than the preset level, the image integrity level is higher than the preset level, and the image type is a specified category, the image clarity level, the image integrity level, and the image attribute type in the image quality result are substituted into the condition, and if the condition is met, it is determined that the target image needs to be subjected to character recognition.
According to the method for detecting the image quality provided by the embodiment of the disclosure, the confidence coefficient that the image output by the whole image definition evaluation model belongs to the definition category can be utilized to determine the whole image definition score, and the field definition score is obtained based on the definition score of each field output by the field definition evaluation model and is jointly used as the definition parameter of the target image, so that the accuracy of the definition parameter is improved. And the execution main body can also determine a plurality of target fields with higher confidence degrees from the target image based on the character detection model to form a target field set, so that the accuracy of the target field set is improved. And the integrity parameter can be determined and obtained by identifying the key point of the target image through the integrity evaluation model, so that the accuracy of integrity judgment is improved. And whether follow-up character recognition is carried out or not can be determined based on the image quality result, so that the character recognition accuracy is improved.
With further reference to fig. 5, as an implementation of the methods shown in the above-mentioned figures, the present disclosure provides an embodiment of an apparatus for detecting image quality, which corresponds to the method embodiment shown in fig. 2, and which is particularly applicable to electronic devices such as terminal devices.
As shown in fig. 5, the apparatus 500 for detecting image quality of the present embodiment includes: an image acquisition unit 501, a model determination unit 502, a parameter determination unit 503, and a quality evaluation unit 504.
An image acquisition unit 501 configured to acquire a target image.
A model determination unit 502 configured to determine at least one model corresponding to the target image.
A parameter determining unit 503 configured to determine an image parameter corresponding to the target image based on the target image and the at least one model; the image parameters include at least one of: a sharpness parameter, an integrity parameter, a type parameter.
A quality evaluation unit 504 configured to determine an image quality result of the target image based on the image parameters.
In some optional implementations of this embodiment, the at least one model includes a whole-image sharpness evaluation model; the definition parameter at least comprises a definition score of the whole image; and, the parameter determination unit 503 is further configured to: determining the confidence coefficient that the target image belongs to the definition category based on the target image and the definition evaluation model of the whole image; based on the confidence, the overall graph sharpness score is determined.
In some optional implementations of this embodiment, the at least one model includes a field sharpness evaluation model; the sharpness parameter comprises at least a field sharpness score; and, the parameter determination unit 503 is further configured to: determining a target field set corresponding to a target image; determining definition scores corresponding to all target fields in the target field set based on the target field set and the field definition evaluation model; and determining field definition scores based on the definition scores corresponding to the target fields.
In some optional implementations of this embodiment, the parameter determining unit 503 is further configured to: determining each candidate field corresponding to the target image and the confidence coefficient of each candidate field based on the target image and a preset character detection model; and determining a target field from each candidate field based on the confidence of each candidate field to obtain a target field set.
In some optional implementations of this embodiment, the parameter determining unit 503 is further configured to: and determining field definition scores based on the definition scores corresponding to the target fields and the confidence degrees of the target fields.
In some optional implementations of this embodiment, the at least one model includes an integrity assessment model; and the parameter determination unit is further configured to: determining each key point corresponding to the target image based on the target image and the integrity evaluation model; and determining the integrity parameter of the target image based on the key points.
In some optional implementations of this embodiment, the apparatus further includes: a category determination unit, further configured to determine a character recognition category corresponding to the target image based on the image quality result; the character recognition category includes performing character recognition or not performing character recognition.
It should be understood that the units 501 to 504 recited in the apparatus 500 for detecting image quality correspond to respective steps in the method described with reference to fig. 2, respectively. Thus, the operations and features described above for the method of using a car phone are equally applicable to the apparatus 500 and the units included therein and will not be described in detail here.
The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.
FIG. 6 illustrates a schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 6, the apparatus 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)602 or a computer program loaded from a storage unit 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data required for the operation of the device 600 can also be stored. The calculation unit 601, the ROM 602, and the RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
A number of components in the device 600 are connected to the I/O interface 605, including: an input unit 606 such as a keyboard, a mouse, or the like; an output unit 607 such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, optical disk, or the like; and a communication unit 609 such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The computing unit 601 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 601 performs the respective methods and processes described above, such as a method for detecting image quality. For example, in some embodiments, the method for detecting image quality may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 600 via the ROM 602 and/or the communication unit 609. When the computer program is loaded into the RAM603 and executed by the computing unit 601, one or more steps of the method for detecting image quality described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured by any other suitable means (e.g. by means of firmware) to perform the method for detecting image quality.
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.
The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.
Claims (17)
1. A method for detecting image quality, comprising:
acquiring a target image;
determining at least one model corresponding to the target image;
determining image parameters corresponding to the target image based on the target image and the at least one model; the image parameters include at least one of: a definition parameter, an integrity parameter, a type parameter;
determining an image quality result of the target image based on the image parameters.
2. The method of claim 1, wherein the at least one model comprises a whole-image sharpness evaluation model; the definition parameters at least comprise definition scores of the whole image; and
the determining, based on the target image and the at least one model, image parameters corresponding to the target image includes:
determining a confidence degree that the target image belongs to a definition category based on the target image and the whole image definition evaluation model;
determining the overall graph sharpness score based on the confidence.
3. The method of claim 1, wherein the at least one model comprises a field sharpness evaluation model; the sharpness parameters include at least a field sharpness score; and
the determining, based on the target image and the at least one model, image parameters corresponding to the target image includes:
determining a target field set corresponding to the target image;
determining definition scores corresponding to all target fields in the target field set based on the target field set and the field definition evaluation model;
and determining the field definition score based on the definition score corresponding to each target field.
4. The method of claim 3, wherein the determining a set of target fields to which the target image corresponds comprises:
determining each candidate field corresponding to the target image and the confidence coefficient of each candidate field based on the target image and a preset character detection model;
and determining a target field from each candidate field based on the confidence of each candidate field to obtain the target field set.
5. The method of claim 3, wherein said determining the field clarity score based on the clarity score corresponding to each target field comprises:
and determining the field definition score based on the definition score corresponding to each target field and the confidence of each target field.
6. The method of claim 1, wherein the at least one model comprises an integrity assessment model; and
the determining, based on the target image and the at least one model, image parameters corresponding to the target image includes:
determining each key point corresponding to the target image based on the target image and the integrity evaluation model;
and determining the integrity parameter of the target image based on the key points.
7. The method of claim 1, further comprising:
determining a character recognition category corresponding to the target image based on the image quality result; the character recognition category includes performing character recognition or not performing character recognition.
8. An apparatus for detecting image quality, comprising:
an image acquisition unit configured to acquire a target image;
a model determination unit configured to determine at least one model corresponding to the target image;
a parameter determination unit configured to determine an image parameter corresponding to the target image based on the target image and the at least one model; the image parameters include at least one of: a definition parameter, an integrity parameter, a type parameter;
a quality evaluation unit configured to determine an image quality result of the target image based on the image parameter.
9. The apparatus of claim 8, wherein the at least one model comprises a full image sharpness evaluation model; the definition parameters at least comprise definition scores of the whole image; and
the parameter determination unit is further configured to:
determining a confidence degree that the target image belongs to a definition category based on the target image and the whole image definition evaluation model;
determining the overall graph sharpness score based on the confidence.
10. The apparatus of claim 8, wherein the at least one model comprises a field sharpness evaluation model; the sharpness parameters include at least a field sharpness score; and
the parameter determination unit is further configured to:
determining a target field set corresponding to the target image;
determining definition scores corresponding to all target fields in the target field set based on the target field set and the field definition evaluation model;
and determining the field definition score based on the definition score corresponding to each target field.
11. The apparatus of claim 10, wherein the parameter determination unit is further configured to:
determining each candidate field corresponding to the target image and the confidence coefficient of each candidate field based on the target image and a preset character detection model;
and determining a target field from each candidate field based on the confidence of each candidate field to obtain the target field set.
12. The apparatus of claim 10, wherein the parameter determination unit is further configured to:
and determining the field definition score based on the definition score corresponding to each target field and the confidence of each target field.
13. The apparatus of claim 8, wherein the at least one model comprises an integrity assessment model; and
the parameter determination unit is further configured to:
determining each key point corresponding to the target image based on the target image and the integrity evaluation model;
and determining the integrity parameter of the target image based on the key points.
14. The apparatus of claim 1, further comprising:
a category determination unit, further configured to determine a character recognition category corresponding to the target image based on the image quality result; the character recognition category includes performing character recognition or not performing character recognition.
15. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
16. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-7.
17. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110930170.XA CN113643260A (en) | 2021-08-13 | 2021-08-13 | Method, apparatus, device, medium and product for detecting image quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110930170.XA CN113643260A (en) | 2021-08-13 | 2021-08-13 | Method, apparatus, device, medium and product for detecting image quality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113643260A true CN113643260A (en) | 2021-11-12 |
Family
ID=78421716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110930170.XA Pending CN113643260A (en) | 2021-08-13 | 2021-08-13 | Method, apparatus, device, medium and product for detecting image quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113643260A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114299037A (en) * | 2021-12-30 | 2022-04-08 | 广州极飞科技股份有限公司 | Method and device for evaluating quality of object detection result, electronic equipment and computer readable storage medium |
| CN115049954A (en) * | 2022-05-09 | 2022-09-13 | 北京百度网讯科技有限公司 | Target identification method, device, electronic equipment and medium |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001291055A (en) * | 2000-04-05 | 2001-10-19 | Casio Comput Co Ltd | Slip recognizing device and medium with its program recorded |
| WO2017060850A1 (en) * | 2015-10-07 | 2017-04-13 | Way2Vat Ltd. | System and methods of an expense management system based upon business document analysis |
| CN108305240A (en) * | 2017-05-22 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Picture quality detection method and device |
| CN108573202A (en) * | 2017-03-17 | 2018-09-25 | 北京旷视科技有限公司 | Identity identifying method, device and system and terminal, server and storage medium |
| CN109345522A (en) * | 2018-09-25 | 2019-02-15 | 北京市商汤科技开发有限公司 | A kind of picture quality screening technique and device, equipment and storage medium |
| US20190172193A1 (en) * | 2017-12-06 | 2019-06-06 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for evaluating image definition, computer device and storage medium |
| CN110766340A (en) * | 2019-10-31 | 2020-02-07 | 支付宝(杭州)信息技术有限公司 | Business auditing method, device and equipment |
| CN111291794A (en) * | 2020-01-21 | 2020-06-16 | 上海眼控科技股份有限公司 | Character recognition method, character recognition device, computer equipment and computer-readable storage medium |
| CN111444795A (en) * | 2020-03-13 | 2020-07-24 | 安诚迈科(北京)信息技术有限公司 | Bill data identification method, electronic device, storage medium and device |
| CN111461070A (en) * | 2020-04-29 | 2020-07-28 | Oppo广东移动通信有限公司 | Text recognition method and device, electronic equipment and storage medium |
| CN111598099A (en) * | 2020-05-18 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Method and device for testing image text recognition performance, testing equipment and medium |
| CN111861991A (en) * | 2020-06-11 | 2020-10-30 | 北京百度网讯科技有限公司 | Method and device for calculating image definition |
| CN111915567A (en) * | 2020-07-06 | 2020-11-10 | 浙江大华技术股份有限公司 | Image quality evaluation method, device, equipment and medium |
| CN112418009A (en) * | 2020-11-06 | 2021-02-26 | 中保车服科技服务股份有限公司 | Image quality detection method, terminal device and storage medium |
| CN112633384A (en) * | 2020-12-25 | 2021-04-09 | 北京百度网讯科技有限公司 | Object identification method and device based on image identification model and electronic equipment |
| CN112668640A (en) * | 2020-12-28 | 2021-04-16 | 泰康保险集团股份有限公司 | Text image quality evaluation method, device, equipment and medium |
| CN112990035A (en) * | 2021-03-23 | 2021-06-18 | 北京百度网讯科技有限公司 | Text recognition method, device, equipment and storage medium |
| CN113052063A (en) * | 2021-03-23 | 2021-06-29 | 北京百度网讯科技有限公司 | Confidence threshold selection method, device, equipment and storage medium |
-
2021
- 2021-08-13 CN CN202110930170.XA patent/CN113643260A/en active Pending
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2001291055A (en) * | 2000-04-05 | 2001-10-19 | Casio Comput Co Ltd | Slip recognizing device and medium with its program recorded |
| WO2017060850A1 (en) * | 2015-10-07 | 2017-04-13 | Way2Vat Ltd. | System and methods of an expense management system based upon business document analysis |
| CN108573202A (en) * | 2017-03-17 | 2018-09-25 | 北京旷视科技有限公司 | Identity identifying method, device and system and terminal, server and storage medium |
| CN108305240A (en) * | 2017-05-22 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Picture quality detection method and device |
| US20190172193A1 (en) * | 2017-12-06 | 2019-06-06 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and apparatus for evaluating image definition, computer device and storage medium |
| CN109345522A (en) * | 2018-09-25 | 2019-02-15 | 北京市商汤科技开发有限公司 | A kind of picture quality screening technique and device, equipment and storage medium |
| CN110766340A (en) * | 2019-10-31 | 2020-02-07 | 支付宝(杭州)信息技术有限公司 | Business auditing method, device and equipment |
| CN111291794A (en) * | 2020-01-21 | 2020-06-16 | 上海眼控科技股份有限公司 | Character recognition method, character recognition device, computer equipment and computer-readable storage medium |
| CN111444795A (en) * | 2020-03-13 | 2020-07-24 | 安诚迈科(北京)信息技术有限公司 | Bill data identification method, electronic device, storage medium and device |
| CN111461070A (en) * | 2020-04-29 | 2020-07-28 | Oppo广东移动通信有限公司 | Text recognition method and device, electronic equipment and storage medium |
| CN111598099A (en) * | 2020-05-18 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Method and device for testing image text recognition performance, testing equipment and medium |
| CN111861991A (en) * | 2020-06-11 | 2020-10-30 | 北京百度网讯科技有限公司 | Method and device for calculating image definition |
| CN111915567A (en) * | 2020-07-06 | 2020-11-10 | 浙江大华技术股份有限公司 | Image quality evaluation method, device, equipment and medium |
| CN112418009A (en) * | 2020-11-06 | 2021-02-26 | 中保车服科技服务股份有限公司 | Image quality detection method, terminal device and storage medium |
| CN112633384A (en) * | 2020-12-25 | 2021-04-09 | 北京百度网讯科技有限公司 | Object identification method and device based on image identification model and electronic equipment |
| CN112668640A (en) * | 2020-12-28 | 2021-04-16 | 泰康保险集团股份有限公司 | Text image quality evaluation method, device, equipment and medium |
| CN112990035A (en) * | 2021-03-23 | 2021-06-18 | 北京百度网讯科技有限公司 | Text recognition method, device, equipment and storage medium |
| CN113052063A (en) * | 2021-03-23 | 2021-06-29 | 北京百度网讯科技有限公司 | Confidence threshold selection method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 赵静;王弦;王奔;蒋国平;谢非;徐丰羽;: "基于神经网络的多类别目标识别", 控制与决策, no. 08, pages 248 - 252 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114299037A (en) * | 2021-12-30 | 2022-04-08 | 广州极飞科技股份有限公司 | Method and device for evaluating quality of object detection result, electronic equipment and computer readable storage medium |
| CN114299037B (en) * | 2021-12-30 | 2023-09-01 | 广州极飞科技股份有限公司 | Quality evaluation method and device for object detection result, electronic equipment and computer readable storage medium |
| CN115049954A (en) * | 2022-05-09 | 2022-09-13 | 北京百度网讯科技有限公司 | Target identification method, device, electronic equipment and medium |
| CN115049954B (en) * | 2022-05-09 | 2023-09-22 | 北京百度网讯科技有限公司 | Target identification method, device, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108229322B (en) | Video-based face recognition method and device, electronic equipment and storage medium | |
| CN112949767B (en) | Sample image increment, image detection model training and image detection method | |
| CN113436100B (en) | Method, apparatus, device, medium, and article for repairing video | |
| CN113065614B (en) | Training method of classification model and method for classifying target object | |
| CN113643260A (en) | Method, apparatus, device, medium and product for detecting image quality | |
| CN112861885A (en) | Image recognition method and device, electronic equipment and storage medium | |
| CN116403083A (en) | Image processing method and device, electronic equipment and storage medium | |
| CN114882321A (en) | Deep learning model training method, target object detection method and device | |
| CN113627361B (en) | Training method and device for face recognition model and computer program product | |
| CN109064464B (en) | Method and device for detecting burrs of battery pole piece | |
| CN113239807B (en) | Method and device for training bill identification model and bill identification | |
| CN112508005B (en) | Method, apparatus, device and storage medium for processing image | |
| CN113902899A (en) | Training method, target detection method, device, electronic device and storage medium | |
| CN114724144B (en) | Text recognition method, training device, training equipment and training medium for model | |
| CN115019057A (en) | Image feature extraction model determining method and device and image identification method and device | |
| CN114093006A (en) | Training method, device and equipment of living human face detection model and storage medium | |
| CN114581711A (en) | Target object detection method, apparatus, device, storage medium, and program product | |
| CN114677566A (en) | Deep learning model training method, object recognition method and device | |
| CN113869253A (en) | Living body detection method, living body training device, electronic apparatus, and medium | |
| CN113903071A (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN113610856A (en) | Method and device for training image segmentation model and image segmentation | |
| CN113379592A (en) | Method and device for processing sensitive area in picture and electronic equipment | |
| CN114494818B (en) | Image processing method, model training method, related device and electronic equipment | |
| CN114092739B (en) | Image processing method, apparatus, device, storage medium, and program product | |
| CN116310592A (en) | Image recognition method, training device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |