WO2021049721A1

WO2021049721A1 - Method and device for performing machine learning and testing on plurality of images

Info

Publication number: WO2021049721A1
Application number: PCT/KR2020/000251
Authority: WO
Inventors: 추연학; 정재호; 박주영
Original assignee: 라온피플 주식회사
Priority date: 2019-09-10
Filing date: 2020-01-07
Publication date: 2021-03-18
Also published as: CN113906471A; KR20210030576A; KR102357729B1

Abstract

A method for performing machine learning and testing on a plurality of images comprises: a step for receiving a plurality of images; and a step for performing both learning and testing on the plurality of images, wherein the testing is performed by reflecting the result of performing the learning.

Description

Method and apparatus for performing machine learning and testing on a plurality of images

The embodiments disclosed herein relate to a method and apparatus for performing machine learning and testing on a plurality of images, and in particular, performing both machine learning and testing on the same image, and reflecting the results of machine learning. It relates to a method and apparatus for performing a test.

2019 project information and resignation

1. Assignment identification number: 1711094483

2. Remarks: This study was conducted as a result of the research of the Ministry of Science and ICT and the Center for Promotion of Information and Communication Technology (IITP-2019-0-01423-001).

Recently, AI technology has been used in the field of machine vision, making a great contribution to factory automation. However, in order to use AI technology in the field of machine vision, it is necessary to determine the existence and degree of defects in the product or to determine the type of product. Machine learning needs to be performed on. For example, machine learning may be performed on a plurality of images photographing good or defective products, or machine learning may be performed on a plurality of images photographing various types of goods.

In a commonly used machine vision technology, learning is first performed using images separately prepared for training in order to perform a test to distinguish between good and defective products through photographed images. For example, by performing learning on a plurality of images of good or defective products, after grasping the characteristics of images of good or defective products, testing the image of the goods to determine whether the goods are good or not It is a method of determining whether the product is defective.

However, in the case of the existing method of separating learning and testing as described above, the following problems exist.

First, not only is it not easy to secure images of defective products due to the characteristics of the production process where the rate of occurrence of defective products is extremely low, and the task of classifying multiple images into images of good products and images of defective products before the learning is performed. It must be done separately, which takes time and effort.

Second, since the test is performed based on the result of one-time learning, the accuracy of the test result is degraded if there is a change in the environment, such as a change in the brightness or position of the light illuminating the article after learning.

In a related art document, Korean Patent Registration No. 10-1867475 discloses a description of clustering similar models by performing unsupervised learning on static data using an autoencoder.

On the other hand, the above-described background technology is technical information that the inventor possessed for derivation of the present invention or acquired during the derivation process of the present invention, and is not necessarily a known technology disclosed to the general public before filing the present invention. .

Embodiments disclosed herein are to provide a machine learning and test execution method and apparatus for increasing test accuracy by reflecting a learning result in real time by performing both machine learning and testing on a plurality of images.

In order to increase the accuracy of machine learning and testing, both learning and testing are performed on a plurality of images, but the test is performed by reflecting the results of learning.

According to any one of the above-described task solving means, it is possible to perform a test by reflecting the learning result in real time by performing both learning and testing on a plurality of images but reflecting the learning result, and thus, the environment It can be expected to increase test accuracy by adaptively responding to changes.

In addition, since there is no need to prepare separate images for training, it is possible to reduce the time and effort required to secure images of defective products or classify images.

The effects that can be obtained in the disclosed embodiments are not limited to the above-mentioned effects, and other effects not mentioned are obvious to those of ordinary skill in the art to which the embodiments disclosed from the following description belong. It will be understandable.

1 is a diagram illustrating a system for performing machine learning and testing on a plurality of images according to an exemplary embodiment.

FIG. 2 is a block diagram illustrating a configuration of the computing device illustrated in FIG. 1.

3 to 8 are flowcharts illustrating a method for performing machine learning and testing on a plurality of images according to embodiments.

According to an embodiment as a technical means for achieving the above-described technical problem, a method of performing machine learning and testing on a plurality of images includes the steps of receiving a plurality of images, and learning and learning about the plurality of images. Performing all tests, but may include performing the test by reflecting the result of performing the learning.

According to another embodiment, as a computer program for performing a method of performing machine learning and testing on a plurality of images in a computing device, the method of performing machine learning and testing on a plurality of images includes: And performing both learning and testing on the plurality of images, and performing the test by reflecting a result of performing the learning.

According to another embodiment, as a computer-readable recording medium in which a program for performing a method of performing machine learning and testing on a plurality of images in a computing device is recorded, machine learning and testing are performed on a plurality of images. The performing method may include receiving a plurality of images and performing both learning and testing on the plurality of images, and performing the test by reflecting the result of performing the learning.

According to another embodiment, an apparatus for performing machine learning and testing includes an input/output unit for receiving operation and data input related to machine learning and testing, displaying data processing results, and communication for transmitting and receiving data with an external device. And a communication unit for performing machine learning and a storage unit for storing a program for performing the test, and a control unit for performing machine learning and testing on a plurality of images by executing the program, wherein the control unit includes: Both learning and testing may be performed on a plurality of images to be received, but the test may be performed by reflecting the result of performing the learning.

Hereinafter, various embodiments will be described in detail with reference to the accompanying drawings. The embodiments described below may be modified and implemented in various different forms. In order to more clearly describe the features of the embodiments, detailed descriptions of matters widely known to those of ordinary skill in the art to which the following embodiments pertain are omitted. In addition, parts not related to the description of the embodiments are omitted in the drawings, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a component is said to be "connected" with another component, this includes not only the case where it is'directly connected' but also the case where it is'connected with another component in the middle.' In addition, when a certain configuration "includes" a certain configuration, it means that other configurations may be further included rather than excluding other configurations unless otherwise specified.

1 is a diagram illustrating a system for performing machine learning and testing on a plurality of images according to an exemplary embodiment. Referring to FIG. 1, a system for performing machine learning and testing on a plurality of images according to an exemplary embodiment may include a photographing unit 10 and a computing device 100.

The system shown in FIG. 1 is a machine vision system for determining whether articles 1 are good or defective, and the photographing unit 10 captures an image of the article 1 and transmits it to the computing device 100 Then, the computing device 100 determines whether the photographed article 1 is good or defective by analyzing the received image. To this end, the computing device 100 learns the received images (hereinafter, it is used in the same meaning as'machine learning') to determine the characteristics of a normal image (an image in which a good product was photographed) and an abnormal image (an image in which a defective product was photographed). I can grasp it.

The photographing unit 10 is a component for photographing the appearance of the article 1 and may be implemented as a camera including an image sensor. In FIG. 1, the photographing unit 10 is illustrated as being a separate component from the computing device 100, but differently, the photographing unit 10 may be a component included in the computing device 100. In addition, the photographing unit 10 may include an illumination for illuminating the article 1 when photographing in order to secure a clearer image.

The photographing unit 10 may photograph items 1 moving through a conveyor belt or the like one by one and transmit the photographed images to the computing device 100.

The computing device 100 performs machine learning and tests on a plurality of images received from the photographing unit 10. The computing device 100 determines whether the product 1 included in the image is a good product or a defective product by analyzing the image in which the product 1 is photographed.

In particular, the computing device 100 performs both learning and testing on a plurality of images, but by reflecting the learning result and performing the test, the learning result may be reflected in the test in real time. That is, the computing device 100 may perform both learning and testing on the same image without separating the training image and the test image from each other. Therefore, it is not necessary to separately prepare an image for learning, so an effect of improving user convenience can be expected. In addition, the test is performed while learning the image that has been tested at the same time, and the result of the learning is reflected when the test is performed on the next image, so even if the surrounding environment such as the brightness or location of the light changes, it is adaptively. In response, it can be expected to increase test accuracy.

A detailed method for the computing device 100 to learn and test a plurality of images will be described below with reference to other drawings.

FIG. 2 is a block diagram illustrating the configuration of the computing device 100 illustrated in FIG. 1. Referring to FIG. 2, the computing device 100 according to an embodiment may include an input/output unit 110, a communication unit 120, a control unit 130, and a storage unit 140.

The input/output unit 110 is a component for receiving inputs such as settings related to machine learning and test performance from a user, and outputting a test result. According to an embodiment, the input/output unit 110 may receive a setting input from a user through input devices such as a keyboard and a mouse, and display results of learning and testing on a screen.

The communication unit 120 is a component for performing communication for transmitting and receiving data with an external device, and may support various types of wired/wireless communication. For example, the communication unit 120 may sequentially receive images photographed of the article 1 from the photographing unit 10. The communication unit 120 may be implemented as a communication chipset supporting various communication protocols.

The controller 130 is a component including at least one processor such as a CPU, and controls the overall operation of the computing device 100. In particular, the control unit 130 may perform machine learning and testing on a plurality of images by executing a program stored in the storage unit 140. A detailed operation of the controller 130 performing machine learning and testing on a plurality of images will be described in detail below.

Various types of programs and data may be stored in the storage unit 140. In particular, a program for performing machine learning and testing on a plurality of images may be stored in the storage unit 140, and a plurality of images to be learned and tested may also be stored.

According to an embodiment, the controller 130 may learn and test a plurality of images received through the communication unit 120, but may perform a test by reflecting a result of performing the learning. In this case, the communication unit 120 may receive images photographed by the article 1 one by one from the photographing unit 10 in order, and the controller 130 may perform learning and testing one by one in the order in which the images were received.

The controller 130 may learn and test each of the plurality of images one by one, but may perform a test on the next image by reflecting the results learned before. For example, when learning and testing on any one image (first image), the controller 130 reflects the learning result for the corresponding image (first image), Perform the test. In addition, the controller 130 may perform learning on the next image (second image) and reflect the result when performing a test on the next image (third image).

In order for the controller 130 to perform a test on an image, it is necessary to know in advance the criteria for testing (hereinafter, referred to as “test criteria”), for example, characteristics of a normal image (an image in which a good product was photographed). These test criteria may be pre-stored in the storage unit 140, or the control unit 130 may directly grasp and store them in the storage unit 140 in the process of learning the images.

Hereinafter, an embodiment in which the control unit 130 directly grasps the test criterion in the process of learning the images will be described.

The controller 130 starts learning one by one for a plurality of images in order. Initially, since the test criterion was not prepared, the controller 130 only learns the image. When learning is performed on images one by one in order, and when a preset certain criterion (hereinafter, referred to as'preparation criterion') is satisfied, the controller 130 determines that preparation for setting the test criterion is completed, and the subsequent images Regarding, both learning and testing are performed, but the test is performed by reflecting the learning results performed until the preparation criteria are satisfied and the learning results of the images in subsequent steps.

In detail, if it is determined that the preparation criterion is satisfied, the controller 130 sets a test criterion based on the learning result performed until the preparation criterion is satisfied, and then performs a test on the subsequent images. At this time, the control unit 130 continuously updates the test criteria based on the result while simultaneously performing the test and learning on the subsequent images. That is, in performing a test on images in a subsequent sequence, the controller 130 performs the test by reflecting the learning results of other images performed just before the test execution.

A method for the controller 130 to determine whether the preparation criterion is satisfied is as follows.

The controller 130 may determine whether or not the preparation criterion is satisfied based on the time at which learning has been performed to the present or the number of images that have been trained to date. This is because test criteria such as characteristics of normal images can be grasped only when learning is performed for at least a predetermined time or on at least a predetermined number of images.

For example, the control unit 130 may check the time at which learning has been performed so far, and determine that the preparation criterion is satisfied if the checked time exceeds a preset value. Alternatively, for example, the controller 130 may check the number of images that have been trained so far, and determine that the preparation criterion is satisfied if the confirmed number exceeds a preset value.

In addition, the controller 130 may determine whether preparations for setting the test criteria are completed in various ways (whether the preparation criteria are satisfied).

When the preparation criterion is satisfied, a method in which the controller 130 performs both learning and testing on images in subsequent steps is as follows.

If it is determined that the preparation criterion is satisfied, the controller 130 determines the characteristics of the normal image based on the learning result performed until the preparation criterion is satisfied, and sets this as a test criterion. Here, the test criterion is described as an example as a characteristic of a normal image, but differently, the characteristic of an abnormal image may be a test criterion.

The controller 130 performs a test on any one image (first image) on the basis of the identified characteristics of the normal image. That is, the controller 130 determines whether the first image is a normal image or an abnormal image. When it is determined that the first image is an abnormal image, the controller 130 may display a guide indicating that a defective product has been detected through the input/output unit 110.

The controller 130 performs learning as well as testing on the first image. The controller 130 updates the characteristics of the normal image based on the result of learning the first image. Therefore, when performing a test on an image (second image) in a sequence following the first image, the controller 130 determines whether the second image is a normal image based on the characteristics of the normal image in which the learning result for the first image is reflected. Can be judged. In this way, the controller 130 may continuously update the test criteria by performing a test on each image and learning at the same time.

On the other hand, in the above, it has been described that the control unit 130 updates the test criterion (characteristic of a normal image) every time it learns about one image. May not be efficient at Therefore, instead of updating the test criterion for each image, the test criterion may be updated only when a preset certain criterion (hereinafter, referred to as “update criterion”) is satisfied. Hereinafter, such an embodiment will be described.

When it is determined that the preparation criterion is satisfied, the controller 130 determines the characteristics of the normal image based on the previously performed learning result. Here, the test criterion is described as an example as a characteristic of a normal image, but differently, the characteristic of an abnormal image may be a test criterion.

The controller 130 performs learning as well as testing on the first image. When the learning of the first image is completed, the controller 130 determines whether or not a preset update criterion is satisfied. In detail, the controller 130 checks the time when the feature of the normal image has been identified or the time when the feature of the normal image has been updated has been completed, and if the time has exceeded a preset value, the update criterion It can be judged as satisfying. Alternatively, the controller 130 checks the number of images that have been trained after the time when the feature of the normal image has been identified or the time when the feature of the normal image has been updated is completed, and if the confirmed number exceeds a preset value, it is updated. It can be determined that the criteria are satisfied.

In this way, the controller 130 reflects the learning result and updates the test criterion only when a certain criterion (update criterion) is satisfied, thereby reflecting the learning result in real time and increasing efficiency.

Hereinafter, a method of performing machine learning and testing on a plurality of images according to embodiments will be described with reference to FIGS. 3 to 8. The method of performing machine learning and testing according to the embodiments illustrated in FIGS. 3 to 8 includes steps processed in a time series by the computing device 100 illustrated in FIGS. 1 and 2. Therefore, even though the contents are omitted below, the contents described above with respect to the computing device 100 shown in FIGS. 1 and 2 are also included in the method of performing machine learning and testing according to the embodiments shown in FIGS. 3 to 8. Can be applied.

3 is a flowchart illustrating a method for performing machine learning and testing on a plurality of images according to an exemplary embodiment.

Referring to FIG. 3, in step 301, the computing device 100 receives a plurality of images. In this case, the computing device 100 may receive a plurality of images at once, or may receive target images one by one while performing learning and testing.

In step 302, the computing device 100 performs both learning and testing on a plurality of images, but performs a test by reflecting the learning result. In other words, the computing device 100 performs learning and testing on each of the plurality of images in order, reflecting the result of learning on the previous image, and performs a test on the next image.

4 is a flowchart illustrating detailed steps included in step 302 of FIG. 3. Referring to FIG. 4, in step 401, the computing device 100 learns a plurality of received images one by one.

In step 402, the computing device 100 determines whether the next image to be learned exists, ends the process if the next image does not exist, and proceeds to step 403 if the next image exists.

In step 403, the computing device 100 determines whether a predetermined criterion (preparation criterion) is satisfied, and if the preparatory criterion is not satisfied, returns to step 401, and if the preparatory criterion is satisfied, the computing device 100 proceeds to step 404. Here, the'preparation criterion' is a criterion for determining whether preparation for setting a criterion (test criterion) for testing is completed, as described above. That is, when the preparation criterion is satisfied, the computing device 100 determines that preparation for setting the test criterion is completed. The computing device 100 determines whether or not the preparation criterion is satisfied based on the time at which the training has been performed to the present or the number of images that have been trained to date, a detailed process will be described with reference to FIGS. 7 and 8 below. .

In step 404, the computing device 100 performs both learning and testing on the remaining images, but performs a test by reflecting the learning results of other images.

When it is determined that the preparation criterion is satisfied, the computing device 100 performs both learning and testing for the subsequent images, but reflects the learning result performed until the preparation criterion is satisfied and the learning results for the subsequent images. To perform the test. In detail, when it is determined that the preparation criterion is satisfied, the computing device 100 sets a test criterion based on a learning result performed until the preparation criterion is satisfied, and then performs tests on images in a subsequent sequence. In this case, the computing device 100 continuously updates the test criteria based on the result while simultaneously performing the test and learning on the subsequent images. That is, when the computing device 100 performs a test on images in a subsequent sequence, the computing device 100 performs the test by reflecting the learning results of other images performed just before the test.

A detailed process in which the computing device 100 performs a test by reflecting the learning result of another image in step 404 will be described below with reference to FIGS. 5 and 6.

5 is a flowchart illustrating detailed steps included in step 404 of FIG. 4. Referring to FIG. 5, in step 501, the computing device 100 identifies features of a normal image based on a result of performing previous learning.

In step 502, the computing device 100 detects an abnormal image by performing a test on the image based on the characteristics of the normal image. The computing device 100 may determine whether each image is a normal image or an abnormal image by comparing features of the normal image identified in step 501 with each image.

In step 503, the computing device 100 performs learning on the image tested in step 502. In step 504, the computing device 100 updates the features of the normal image based on the result of performing the learning in step 503.

In step 505, the computing device 100 determines whether the next image exists, returns to step 502 if the next image exists, and terminates the process if the next image does not exist.

Meanwhile, as described above, in updating the features of a normal image, the computing device 100 does not update each time learning of one image is completed, and only updates when a predetermined condition (update condition) is satisfied. By doing so, it is possible to increase the efficiency while reflecting the learning result in real time. This embodiment will be described with reference to FIG. 6 below.

6 is a flowchart illustrating detailed steps included in step 404 of FIG. 4. Referring to FIG. 6, in step 601, the computing device 100 identifies features of a normal image based on a result of performing previous learning.

In step 602, the computing device 100 detects an abnormal image by performing a test on the image based on features of the normal image. The computing device 100 may determine whether each image is a normal image or an abnormal image by comparing the features of the normal image identified in step 601 with each image.

In step 603, the computing device 100 learns the image that has been tested in step 602. In step 604, the computing device 100 determines whether there is a next image, and if there is no next image as a result of the determination, the process ends. Conversely, if there is a next image, the process proceeds to step 605.

In step 605, the computing device 100 determines whether or not a preset criterion (update criterion) is satisfied. In detail, the computing device 100 checks the time at which the learning was performed after the time when the feature of the normal image was completed or the time when the update of the feature of the normal image was completed, and if the checked time exceeds a preset value, it is updated. It can be determined that the criteria are satisfied. Alternatively, the computing device 100 checks the number of images that have been trained after the time when the feature identification of the normal image has been completed or the time when the update of the features of the normal image is completed, and if the confirmed number exceeds a preset value, It can be determined that the update criteria are satisfied.

As a result of the determination in step 605, if it is determined that the preset criterion is not satisfied, the computing device 100 returns to and performs step 602 again. Conversely, if it is determined that the preset criterion is satisfied, the computing device 100 proceeds to step 606.

In step 606, the computing device 100 updates the features of the normal image based on the results of learning the images after the features of the normal image are last updated. After performing step 606, the computing device 100 returns and performs step 602 again.

Hereinafter, a detailed method of determining whether the computing device 100 satisfies the preparation criterion in step 403 of FIG. 4 will be described with reference to FIGS. 7 and 8.

Referring to FIG. 7, in step 701, the computing device 100 checks the time at which learning has been performed. In step 702, the computing device 100 determines whether the checked time exceeds a preset value, and if it exceeds the preset value, proceeds to step 404 of FIG. 4, and if it is less than the preset value, step 401 of FIG. Do it again.

Referring to FIG. 8, in step 801, the computing device 100 checks the number of images that have been trained so far. In step 802, the computing device 100 determines whether the confirmed number exceeds a preset value, and if it exceeds the preset value, proceeds to step 404 of FIG. 4, and if it is less than the preset value, step 401 of FIG. 4 Do it again.

In this way, the computing device 100 learns the images until the preparation criterion is satisfied, thereby grasping the criteria for testing, and learning the subsequent images and at the same time grasping the learning results up to the previous image. The test can be performed based on the established test criteria.

According to the above-described embodiments, by performing both learning and testing on a plurality of images, but performing a test by reflecting the learning result, it is possible to perform the test by reflecting the learning result in real time, thus changing the environment. It can be expected to increase test accuracy by adaptively responding to.

The term'~ unit' used in the above embodiments refers to software or hardware components such as field programmable gate array (FPGA) or ASIC, and'~ unit' performs certain roles. However,'~ part' is not limited to software or hardware. The'~ unit' may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example,'~ unit' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and procedures. , Subroutines, segments of program patent code, drivers, firmware, microcode, circuitry, data, database, data structures, tables, arrays, and variables.

The components and functions provided within the'~ units' may be combined into a smaller number of elements and'~ units', or may be separated from the additional elements and'~ units'.

In addition, components and'~ units' may be implemented to play one or more CPUs in a device or a security multimedia card.

The method of performing machine learning and testing according to the embodiments described with reference to FIGS. 3 to 8 may also be implemented in the form of a computer-readable medium that stores instructions and data executable by a computer. In this case, the instructions and data may be stored in the form of a program code, and when executed by a processor, a predetermined program module may be generated to perform a predetermined operation. Further, the computer-readable medium may be any available medium that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. In addition, the computer-readable medium may be a computer recording medium, which is volatile and non-volatile implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. It may include both volatile, removable and non-removable media. For example, the computer recording medium may be a magnetic storage medium such as HDD and SSD, an optical recording medium such as CD, DVD, and Blu-ray disk, or a memory included in a server accessible through a network.

In addition, the method of performing machine learning and testing according to the embodiments described with reference to FIGS. 3 to 8 may be implemented as a computer program (or computer program product) including instructions executable by a computer. The computer program includes programmable machine instructions processed by a processor, and may be implemented in a high-level programming language, an object-oriented programming language, an assembly language, or a machine language. . Further, the computer program may be recorded on a tangible computer-readable recording medium (eg, a memory, a hard disk, a magnetic/optical medium or a solid-state drive (SSD), etc.).

Accordingly, the method of performing machine learning and testing according to the embodiments described with reference to FIGS. 3 to 8 may be implemented by executing the computer program as described above by the computing device. The computing device may include at least some of a processor, a memory, a storage device, a high-speed interface connected to the memory and a high-speed expansion port, and a low-speed interface connected to the low-speed bus and the storage device. Each of these components is connected to each other using a variety of buses, and can be mounted on a common motherboard or in other suitable manner.

Here, the processor can process commands within the computing device. Such commands include, for example, to display graphic information for providing a GUI (Graphic User Interface) on an external input or output device, such as a display connected to a high-speed interface. This could be a command stored in memory or a storage device. As another embodiment, multiple processors and/or multiple buses may be used with multiple memories and memory types as appropriate. In addition, the processor may be implemented as a chipset formed by chips including a plurality of independent analog and/or digital processors.

The memory also stores information within the computing device. As an example, the memory may be composed of volatile memory units or a set of them. As another example, the memory may be composed of a nonvolatile memory unit or a set of them. Also, the memory may be another type of computer-readable medium such as a magnetic or optical disk.

In addition, the storage device may provide a large-capacity storage space to the computing device. The storage device may be a computer-readable medium or a configuration including such a medium, for example, devices in a storage area network (SAN) or other configurations, a floppy disk device, a hard disk device, an optical disk device, Alternatively, it may be a tape device, a flash memory, or another semiconductor memory device or device array similar thereto.

The above-described embodiments are for illustrative purposes only, and those of ordinary skill in the art to which the above-described embodiments belong can easily transform into other specific forms without changing the technical spirit or essential features of the above-described embodiments. You can understand. Therefore, it should be understood that the above-described embodiments are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope to be protected through the present specification is indicated by the claims to be described later rather than the detailed description, and should be interpreted as including all changes or modified forms derived from the meaning and scope of the claims and the concept of equivalents thereof. .

Claims

In a method for performing machine learning and testing on a plurality of images,

Receiving a plurality of images; And

And performing both learning and testing on the plurality of images, and performing the test by reflecting a result of performing the learning.
The method of claim 1,

The step of performing the test,

Determining whether a preset first criterion is satisfied while learning the received images one by one; And

As a result of the determination, if the first criterion is satisfied, both learning and testing are performed for the remaining images, but performing a test by reflecting the learning result performed until the first criterion is satisfied and the learning result for the remaining images. Method comprising the.
The method of claim 2,

The step of performing a test by reflecting the already performed learning result,

Grasping the characteristics of the normal image based on the learning result performed until the first criterion is satisfied;

Determining whether the first image is a normal image based on the identified characteristics of the normal image;

Performing learning on the first image; And

And updating a feature of the normal image based on a result of performing learning on the first image.
The method of claim 2,

The step of performing a test by reflecting the already performed learning result,

Grasping the characteristics of the normal image based on the learning result performed until the first criterion is satisfied;

Determining whether the first image is a normal image based on the identified characteristics of the normal image;

Performing learning on the first image;

Determining whether a predetermined second criterion is satisfied; And

If the determination result satisfies the second criterion, the feature of the normal image is updated based on the result of learning the first image, and if the determination result does not satisfy the second criterion, the feature of the normal image is And performing testing and learning on the next image.
The method of claim 2,

The step of determining whether the first criterion is satisfied,

Checking the time at which the learning has been performed up to now; And

And determining that the first criterion is satisfied if the checked time exceeds a preset time.
The method of claim 2,

The step of determining whether the first criterion is satisfied,

Checking the number of images that have been trained so far; And

And determining that the first criterion is satisfied if the confirmed number exceeds a preset number.
A computer-readable recording medium on which a program for performing the method according to claim 1 is recorded.
A computer program performed by an apparatus for performing machine learning and testing and stored in a medium for performing the method according to claim 1.
In the device for performing machine learning and testing,

An input/output unit for receiving operation and data input related to machine learning and testing, and displaying a data processing result;

A communication unit for performing communication for transmitting and receiving data with an external device;

A storage unit storing programs for machine learning and test execution; And

And a control unit for performing machine learning and testing on a plurality of images by executing the program,

The controller, wherein the controller performs both learning and testing on a plurality of images received through the communication unit, and performing the test by reflecting a result of performing the learning.
The method of claim 9,

In performing the test, the control unit,

While learning the received images one by one, it is determined whether a preset first criterion is satisfied, and if the first criterion is satisfied as a result of the determination, both learning and testing are performed on the remaining images, but the first criterion is satisfied. The apparatus, characterized in that the test is performed by reflecting the learning result performed until time and the learning result of the remaining images.
The method of claim 10,

In the control unit performs a test by reflecting the already performed learning result,

Based on the learning result performed until the first criterion is satisfied, the characteristic of the normal image is identified, based on the identified characteristic of the normal image, it is determined whether the first image is a normal image, and the first image is learned. And updating characteristics of the normal image based on a result of performing learning on the first image.
The method of claim 10,

In the control unit performs a test by reflecting the already performed learning result,

Based on the learning result performed until the first criterion is satisfied, the characteristic of the normal image is identified, based on the identified characteristic of the normal image, it is determined whether the first image is a normal image, and the first image is learned. After performing, it is determined whether a preset second criterion is satisfied, and if the determination result satisfies the second criterion, the characteristic of the normal image is updated based on the result of learning the first image, and the determination result And if the second criterion is not satisfied, tests and learning are performed on the next image based on the characteristics of the normal image.
The method of claim 10,

In determining whether the control unit satisfies the first criterion,

A device, characterized in that, checking a time for performing learning up to now, and determining that the first criterion is satisfied if the checked time exceeds a preset time.
The method of claim 10,

In determining whether the control unit satisfies the first criterion,

A device, characterized in that, checking the number of images that have been trained so far, and determining that the first criterion is satisfied if the confirmed number exceeds a preset number.