Apparatus and method for diagnosing defects using a continuous learning-based learning model

Abstract

To provide a device for diagnosing a defect using a learning model based on continual learning and a method therefor.SOLUTION: In a system including a radiographic device, a scanner, a testing device, and a storage device, a training unit of the testing device 10 is configured to load a buffer training data set, which is training data selected according to a degree of influence on prediction performance of a current learning model from among past training data sets in continual learning, and train the current learning model using the buffer training data set and a current training data set. A detection unit is configured to detect a defect in a radiographic image using the current learning model when the radiographic image is input.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to defect diagnosis techniques, and more particularly to an apparatus and method for diagnosing defects using a continuous learning-based learning model.

[0002] The present invention relates to a training data selection technique, and more particularly to an apparatus for selecting training data for continuous training and a method therefor.

[0003] The present invention relates to a training data update technique, and more particularly to an apparatus for updating training data for continuous training and a method therefor. [Background technology]

[0004] Radiographic testing (RT) is an inspection method that detects defects inside a test object by passing radiation such as X-rays or gamma rays through the test object, depending on the conditions of use and the application, and forming an X-ray film on the test object. It is currently the most widely used non-destructive testing method for detecting internal defects.

[0005] The main task of defect diagnosis based on AI-based industrial radiography (RT) is object detection, and the accuracy of object detection must be maintained over time. When AI models are operated, transfer learning is typically used as a way to re-learn the features of new data sets that are generated over time, but this can lead to forgetting, which reduces accuracy for previous data. If all previous training data sets are retained to prevent this, storage space issues arise. Summary of the Invention [Problem to be solved by the invention]

[0006] SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus and method for diagnosing defects using a continuous learning-based learning model.

[0007] SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus for selecting training data for continuous training and a method therefor.

[0008] SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus for updating training data for continuous training and a method therefor. [Means for solving the problem]

[0009] To achieve the above object, a method for diagnosing defects according to a preferred embodiment of the present invention includes a step in which a learning unit trains a current learning model using a buffer learning data set and a current learning data set pre-stored in a buffer, and a step in which a detection unit detects defects in a radiological image using the current learning model when the radiological image is input.

[0010] The step of training the current training model includes a step in which the training unit loads a buffer training dataset, which is training data selected from past training datasets in continuous training based on the degree of influence it has on the predictive performance of the current training model, and a step in which the current training model is trained using the buffer training dataset and the current training dataset.

[0011] In the method, after the step of training the current training model, a switching unit switches the training cycle of successive training. of Of which, the learning sessions two or more before the current learning model of a step of calculating a degree of change from a plurality of past learning models to the current learning model; a step of the switching unit determining whether the degree of change is equal to or greater than a reference value; a step of the switching unit determining an update to a buffer learning data set if the degree of change is equal to or greater than the reference value; and a step of the update unit determining an update to a buffer learning data set if the degree of change is equal to or greater than the reference value; inWhen a future learning model to be trained is trained using the current learning dataset, the method further includes a step of calculating the degree of influence of each of a plurality of learning data of the current learning dataset on the predictive performance of the future learning model, and a step in which the update unit selects a portion of the plurality of learning data in accordance with the order of the degree of influence and updates the buffer learning dataset.

[0012] The step of calculating the influence includes a step in which the update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set, and a step in which the update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set. in The first past study session in The method includes a step of calculating a stability score indicating the probability that the predicted value of the learned first past learning model differs, and a step in which the update unit calculates a weighted average of the plasticity score and the stability score as the influence.

[0013] The step of calculating the stability score includes: Formula

number

number

number

number

number

[0014] The step of calculating the plasticity score includes: Formula

number

number

number

number

number

[0015] The step of calculating the plasticity score includes: Formula

number

number

[0016] The step of calculating the influence degree includes: Formula

number

[0017] The step of calculating the degree of change is carried out by comparing the weight vector of the current learning model with the weight vector of the previous learning cycle of the current learning model. in a first gradient vector indicating a change between the weight vector of the trained first past learning model and the weight vector of the first past learning model in a previous learning cycle; in The degree of change is derived according to the degree of similarity with a second gradient vector that indicates a change between the weight vector of the trained second past-learning model.

[0018] The similarity is Formula

number

[0019] To achieve the above object, an apparatus for diagnosing defects according to a preferred embodiment of the present invention includes a learning unit that trains a current learning model using a buffer learning data set and a current learning data set pre-stored in a buffer, and a detection unit that detects defects in a radiological image using the current learning model when a radiological image is input.

[0020] The learning unit is characterized in that, in continuous learning, it loads a buffer learning dataset, which is learning data selected from past learning datasets according to the degree of influence it has on the predictive performance of the current learning model, and trains the current learning model using the buffer learning dataset and the current learning dataset.

[0021] The device performs a learning cycle of continuous learning. of Of which, the learning sessions two or more before the current learning model of a switching unit that calculates a degree of change from a plurality of past learning models to the current learning model, determines whether the degree of change is equal to or greater than a reference value, and determines an update to a buffer learning data set if the degree of change is equal to or greater than the reference value; and in and an update unit that, when a future learning model to be trained is trained using the current learning dataset, calculates the degree of influence of each of a plurality of learning data of the current learning dataset on the predictive performance of the future learning model, and selects a portion of the plurality of learning data in accordance with the order of the degree of influence to update the buffer learning dataset.

[0022] The update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set, and calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set. in The first past study session in The method is characterized in that a stability score indicating the probability that the predicted value of the learned first past learning model differs is calculated, and the weighted average of the plasticity score and the stability score is calculated as the influence.

[0023] The update unit Formula

number

number

number

number

number

[0024] The update unit Formula

number

number

number

number

number

[0025] The update unit Formula

number

number

[0026] The update unit Formula

number

[0027] The switching unit selects a weight vector of a current learning model and a weight vector of a previous learning cycle of the current learning model. in a first gradient vector indicating a change between the weight vector of the trained first past learning model and the weight vector of the first past learning model in a previous learning cycle; in The degree of change is derived according to the degree of similarity with a second gradient vector that indicates a change between the weight vector of the trained second past-learning model.

[0028] The switching unit Formula

number

[0029] To achieve the above object, a method for selecting training data according to a preferred embodiment of the present invention includes the steps of: a training unit training a current training model using a buffer training data set and a current training data set previously stored in a buffer; and an update unit selecting a next training cycle of the current training model in continuous training. in When a future learning model to be trained is trained using the current learning dataset, the method includes a step of calculating the degree of influence of each of multiple learning data in the current learning dataset on the predictive performance of the future learning model, and a step in which the update unit selects a portion of the multiple learning data in accordance with the order of the degree of influence and updates the buffer learning dataset.

[0030] The step of calculating the influence includes a step in which the update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set, and a step in which the update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set. in The first past study session in The method includes a step of calculating a stability score indicating the probability that the predicted value of the learned first past learning model differs, and a step in which the update unit calculates a weighted average of the plasticity score and the stability score as the influence.

[0031] The step of calculating the stability score includes the update unit: Formula

number

number

number

number

number

[0032] The step of calculating the plasticity score includes the steps of: Formula

number

number

number

number

number

[0033] The method further includes, before the step of calculating the influence, a step in which the learning unit constructs derived training data, which is training data derived from the current training dataset, and a step in which the learning unit trains a gradient prediction model using the derived training data.

[0034] The step of calculating the plasticity score is characterized in that the switching unit calculates the plasticity score using the gradient prediction model.

[0035] The derived learning data includes a weight vector of a target layer of a current learning model that is corrected each time the learning unit divides the current learning data set into epoch units and iteratively learns the current learning model, and a target gradient vector that indicates a difference between a weight vector from a previous epoch and a weight vector from a current epoch that is corrected each time the learning unit divides the current learning data set into epoch units and iteratively learns the current learning model.

[0036] The step of calculating the plasticity score includes the steps of: Formula

number

number

[0037] The step of calculating the degree of influence is performed by the update unit. Formula

number

[0038] In order to achieve the above object, a device for selecting training data according to a preferred embodiment of the present invention includes a training unit for training a current training model using a buffer training data set and a current training data set pre-stored in a buffer, and a training unit for selecting a next training data set for the current training model in successive training. in and an update unit that, when a future learning model to be trained is trained using the current learning dataset, calculates the degree of influence of each of a plurality of learning data of the current learning dataset on the predictive performance of the future learning model, and selects a portion of the plurality of learning data in accordance with the order of the degree of influence to update the buffer learning dataset.

[0039] The update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set, and calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model will differ for the learning data of the current learning data set. in The first past study session in The method is characterized in that a stability score indicating the probability that the predicted value of the learned first past learning model differs is calculated, and the weighted average of the plasticity score and the stability score is calculated as the influence.

[0040] The update unit is

number

number

number

number

number

[0041] The update unit Formula

number

number

number

number

number

[0042] The learning unit generates derived learning data, which is learning data derived from the current learning data set, and trains a gradient prediction model using the generated derived learning data, and the switching unit calculates the plasticity score using the gradient prediction model.

[0043] The derived learning data includes a weight vector of a target layer of a current learning model that is corrected each time the learning unit divides the current learning data set into epoch units and iteratively learns the current learning model, and a target gradient vector that indicates a difference between a weight vector from a previous epoch and a weight vector from a current epoch that is corrected each time the learning unit divides the current learning data set into epoch units and iteratively learns the current learning model.

[0044] The update unit Formula

number

number

[0045] The update unit Formula

number

[0046] To achieve the above object, a method for updating training data according to a preferred embodiment of the present invention includes: a training unit training a current training model using a buffer training data set and a current training data set previously stored in a buffer; and a switching unit switching the training cycle of successive training. of Of which, the learning sessions two or more before the current learning model ofThe method includes a step of calculating the degree of change from a past learning model to the current learning model, a step in which the switching unit determines whether the degree of change is equal to or greater than a reference value, a step in which the switching unit determines an update to the buffer learning dataset if the degree of change is equal to or greater than the reference value, and a step in which the update unit extracts at least a portion of learning data from the current learning dataset and updates the buffer learning dataset.

[0047] The step of calculating the degree of change may be performed by the switching unit comparing the weight vector of the current learning model with the weight vector of the previous learning cycle of the current learning model. in a first gradient vector indicating a change between the weight vector of the trained first past learning model and the weight vector of the first past learning model in a previous learning cycle; in The degree of change is derived according to the degree of similarity with a second gradient vector that indicates a change between the weight vector of the trained second past-learning model.

[0048] The similarity is Formula

number

[0049] The step of determining whether the degree of change is equal to or greater than a reference value is characterized in that the switching unit determines that the degree of change is equal to or greater than a reference value if the similarity between the first gradient vector and the second gradient vector is less than a reference value.

[0050] The method further includes, after the step of determining whether the degree of change is equal to or greater than a reference value, a step of the switching unit erasing the current training data set if the degree of change is less than the reference value.

[0051] The buffer training data set pre-stored in the buffer is training data selected from past training data sets in successive training according to the degree of influence on the prediction performance of the current training model.

[0052] To achieve the above object, a device for updating training data according to a preferred embodiment of the present invention includes a training unit for training a current training model using a buffer training data set and a current training data set previously stored in a buffer; of Of which, the learning sessions two or more before the current learning model of The system includes a switching unit that calculates a degree of change from a past learning model to the current learning model, determines whether the degree of change is equal to or greater than a reference value, and, if the degree of change is equal to or greater than the reference value, determines an update to the buffer learning data set; and an update unit that extracts at least a portion of learning data from the current learning data set and updates the buffer learning data set.

[0053] The switching unit selects a weight vector of a current learning model and a weight vector of a previous learning cycle of the current learning model. in a first gradient vector indicating a change between the weight vector of the trained first past learning model and the weight vector of the first past learning model in a previous learning cycle; in The degree of change is derived according to the degree of similarity with a second gradient vector that indicates a change between the weight vector of the trained second past-learning model.

[0054] The switching unit Formula

number

[0055] The switching unit determines that the degree of change is equal to or greater than a reference value if the similarity between the first gradient vector and the second gradient vector is less than a reference value.

[0056] The switching unit may delete the current training data set if the degree of change is less than a reference value.

[0057] The buffer training data set pre-stored in the buffer is training data selected from past training data sets in successive training according to the degree of influence on the prediction performance of the current training model. [Effects of the Invention]

[0058] According to the present invention, by updating the buffer learning data set according to the predictive performance of the future learning model, and then continuously learning the learning model using the new learning data set together with the buffer learning data set, it is possible to overcome catastrophic forgetting while minimizing storage space, and continuously provide a learning model that provides high accuracy at minimal cost. [Brief explanation of the drawings]

[0059] [Figure 1] FIG. 1 is a diagram illustrating the configuration of a system for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. [Figure 2] FIG. 1 is a diagram illustrating the configuration of an apparatus for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. [Figure 3] FIG. 10 is a diagram illustrating a learning data set and a learning model according to a learning round in continuous learning according to an embodiment of the present invention. [Figure 4] 1 is a flowchart illustrating a method for selecting training data for continuous training according to an embodiment of the present invention. [Figure 5] 1 is a flowchart illustrating a method for training a gradient prediction model according to an embodiment of the present invention. [Figure 6] 1 is a flowchart illustrating a method for calculating the degree of influence of learning data on the prediction performance of a future learning model according to an embodiment of the present invention. [Figure 7] 1 is a flowchart illustrating a method for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. [Figure 8] 10A-10C are example screen shots illustrating a method for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. [Figure 9] FIG. 1 illustrates a computing device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0060] Although the present invention can be implemented in various forms by adding various modifications, specific embodiments will be illustrated and described in detail in the detailed description, but it should be understood that this is not intended to limit the present invention to the specific embodiments, but rather to include all modifications, equivalents, or alternatives that fall within the spirit and technical scope of the present invention.

[0061] The terms used in the present invention are merely used to describe specific embodiments and are not intended to limit the present invention. A singular expression includes a plural expression unless the context clearly indicates otherwise. It should be understood that, in the present invention, terms such as "comprise" or "have" are intended to specify the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, and do not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0062] In particular, the terms and phrases used in the following specification and claims should not be interpreted as being limited to their ordinary or dictionary meanings, but should be interpreted as meanings and concepts that correspond to the technical idea of ​​the present invention, in accordance with the principle that the inventor can appropriately define the concepts of terms to best describe his / her invention. In particular, in the embodiments of the present invention, estimation or inference means deriving a result of calculation based on what a learning model (LM: Machine Learning Model / Deep Learning Model) has learned.

[0063] First, a system for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention will be described. FIG. 1 is a diagram for explaining the configuration of a system for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. FIG. 2 is a diagram for explaining the configuration of an apparatus for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. FIG. 3 is a diagram for explaining a learning cycle in continuous learning according to an embodiment of the present invention. to FIG. 1 is a diagram for explaining a learning data set and a learning model based on the above.

[0064] Referring to FIG. 1, a system for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention includes a radiography apparatus RTA, a scanner SC, an inspection apparatus 10, and a storage apparatus 40.

[0065] The radiographic imaging apparatus RTA performs radiography on an object to which an image quality indicator (IQI) is attached, and outputs a radiographic inspection film in which the object to which the IQI is attached is photographed. The object may be a pipe, a tube, or the like.

[0066] The scanner SC scans the radiographic inspection film to generate a digitized radiation image. Once the digitized radiation image is generated, it is input to the inspection device 10. At this time, the radiation image may be input directly from the scanner SC or may be stored in another storage medium and then input from the storage medium.

[0067] The inspection device 10 trains a current training model using a buffer training data set and a current training data set previously stored in a buffer BF through continuous training, and detects defects in a radiological image using the current training model, which is the most recent training model among the training models trained through continuous training. The inspection device 10 can then output a report including a radiological image in which defects are detected using a bounding box BB.

[0068] The storage device 40 may be a cloud server or a database server. The buffer BF of the storage device 40 is for storing data, and the buffer BF may store a buffer training data set according to an embodiment of the present invention. Here, the buffer training data set is a set of data from past training rounds based on the current time in continuous training. of The buffer BF is training data selected from the past training data sets used in training according to the degree of influence it has on the prediction performance of the current training model. Meanwhile, although the buffer BF has been described as a storage medium included in the storage device 40, according to another embodiment of the present invention, the buffer BF may be a storage medium included in the inspection device 10.

[0069] Referring to FIG. 2, the inspection device 10 includes a collection unit 100, a learning unit 200, a switching unit 300, an update unit 400, and a detection unit 500.

[0070] The collection unit 100 continuously collects learning data for continuous learning and constructs a current learning data set. After constructing the current learning data set, the collection unit 100 provides the constructed current learning data set to the learning unit 200.

[0071] The learning unit 200 generates a learning model through learning (deep learning or machine learning). The learning model according to an embodiment of the present invention is for detecting an object learned from an image. The learning model according to an embodiment of the present invention detects an object, i.e., a defect, learned from a radiation image. In this case, the learning model can detect an area occupied by a defect in the radiation image using a bounding box BB. Examples of such learning models include CNN, YOLO, RCNN, and Faster RCNN. The learning model includes multiple interconnected layers (or modules), and the multiple layers (or modules) perform multiple operations. The multiple layers (or modules) are connected by weights (W). That is, the output of an operation result of any one layer (or module) is applied with a weight and input to the operation of the next layer. The learning model derives an output by performing multiple operations on input data, to which weights are applied between multiple layers (or modules). In other words, the learning model performs multiple operations connected by weights between multiple layers (or modules). Such multiple operations linked by weights between multiple layers (or modules) of a learning model are referred to as "weight operations."

[0072] In particular, the learning unit 200 according to the embodiment of the present invention learns a learning model through continuous learning.

[0073] Referring to FIG. 3, in an embodiment of the present invention, continuous learning is a process in which new learning data is continuously collected and the continuously collected learning data is continuously used for learning. Go After dividing the data into multiple training data sets, the training data is Go In the embodiment of the present invention, the learning cycle is a continuous learning method for a learning model using a learning data set. teeth The learning cycle can be divided by at least one of time, the number of collected learning data, and a specific event. to A training data set including multiple training data divided by Go Since learning is continuous, the learning model is continuously updated and in Learning models that are different from the derived learning models are continually generated.

[0074] In the embodiment of the present invention, the reference time point is set as the current learning time. t When this is done, the current learning session t The learning model learned in the current learning round will be referred to as the current learning model. Also, the learning dataset used to learn the current learning model will be referred to as the current learning dataset. t Previous lesson in The first past study session t -1 to current learning session t It includes multiple training data collected up to

[0075] Also, the previous learning session of the current learning model in The first past study session t The learning model learned in -1 will be referred to as the first past learning model. Also, the learning dataset for learning the first past learning model will be referred to as the first past learning dataset. The first past learning dataset is the data set from the first past learning round. t -1 previous learning session in A second past study session t -2 to 1st past study session t Contains multiple training data collected up to -1.

[0076] And the current learning session t All previous learning sessions are compared t The learning models for t-1, t-2, ... are called past learning models. t All previous learning sessions are compared t The training data sets of -1, t-2, ... are referred to as past training data sets.

[0077] On the other hand, the next learning cycle of the current learning model in A future learning session t The learning model trained in +1 is called the future learning model. The learning dataset for training the future learning model is called the future learning dataset. The future learning dataset is the dataset for training the current learning session. t Future Learning Session t Contains multiple learning data collected up to +1.

[0078] Referring again to FIG. 2, the switching unit 300 determines whether and when to update the buffer learning data set according to an embodiment of the present invention.

[0079] The switching unit 300 performs the learning cycle of the continuous learning. of Of which, the learning sessions two or more before the current learning model of To this end, the switching unit 300 derives the degree of change from a plurality of past learning models to the current learning model. in The switching unit 300 then derives a first gradient vector that indicates the change between the weight vector of the first past learning model and the weight vector of the first past learning model in the previous learning cycle. inThe switching unit 300 then derives a second gradient vector indicating a change between the weight vector of the trained second past-learned model and the first gradient vector. The switching unit 300 then derives a degree of change according to the similarity between the first gradient vector and the second gradient vector. The switching unit 300 then determines whether the degree of change is equal to or greater than a reference value. That is, if the similarity between the first gradient vector and the second gradient vector is less than the reference value, the switching unit 300 determines that the degree of change is equal to or greater than the reference value. If the similarity is equal to or greater than the reference value, the switching unit 300 determines that the degree of change is equal to or greater than the reference value. less than If the degree of change is equal to or greater than the reference value (the similarity is less than the reference value), the switching unit 300 determines to update the buffer training data set. On the other hand, if the degree of change is less than the reference value (the similarity is equal to or greater than the reference value), the switching unit 300 deletes the current training data set.

[0080] The update unit 400 calculates the degree of influence of each of the plurality of learning data in the current learning data set on the prediction performance of the future learning model. of When a future learning model, which is a learning model, is trained using a current learning dataset, the degree of influence of each of the multiple learning data in the current learning dataset on the predictive performance of the future learning model is calculated.The update unit 400 then selects a portion of the multiple learning data in the current learning dataset in accordance with the calculated order of influence and updates the buffer learning dataset.At this time, the update unit 400 updates the buffer learning dataset by deleting buffer learning data previously stored in the buffer BF and storing learning data selected from the multiple learning data in accordance with the order of influence in the buffer.The update unit 400 also erases the remaining learning data from the multiple learning data that was not selected in accordance with the order of influence.

[0081] Next, a method for selecting training data for continuous training according to an embodiment of the present invention will be described. Figure 4 illustrates a method for selecting training data for continuous training according to an embodiment of the present invention.

[0082] Referring to FIG. 4, the collection unit 100 continuously collects training data to form a current training data set in step S110.

[0083] In step S120, the learning unit 200 loads a buffer training data set pre-stored in the buffer BF, and then generates a current training model by training using the buffer training data set pre-stored in the buffer BF and the current training data set. Here, the buffer training data set pre-stored in the buffer BF is used for the past training rounds in the continuous training. of This is training data selected from past training data sets used in training based on the degree of influence it has on the predictive performance of the current training model.

[0084] In step S130, the switching unit 300 performs the learning cycle of the continuous learning. of Of which, the learning sessions two or more before the current learning model of The degree of change from a plurality of past learning models to the current learning model is derived. Step S130 will be described in more detail as follows.

[0085] First, the switching unit 300 switches the weight vector of the current learning model and the weight vector of the previous learning cycle of the current learning model. in A first gradient vector is derived that indicates the change between the weight vector of the first previously learned model and the weight vector of the previously learned model.

[0086] Next, the switching unit 300 switches the weight vector of the first past learning model and the weight vector of the previous learning cycle of the first past learning model. in The switching unit 300 then derives a second gradient vector indicating a change between the weight vector of the second previously learned model and the first gradient vector, and then derives a change degree according to the similarity between the first gradient vector and the second gradient vector.

[0087] At this time, the switching unit 300 can calculate the similarity using the following equation 1. [Number 1]

number

[0088] Here, CS indicates the similarity. t is the first gradient vector between the weight vector of the current training model and the weight vector of the first past training model. And g t-1 is the second gradient vector between the weight vector of the first past training model and the weight vector of the second past training model. T t-1 is the transpose of the second gradient vector.

[0089] Next, in step S140, the switching unit 300 determines whether the degree of change is equal to or greater than a reference value. At this time, if the similarity between the first gradient vector and the second gradient vector is less than the reference value according to Equation 1, the switching unit 300 determines that the degree of change is equal to or greater than the reference value. On the other hand, if the similarity between the first gradient vector and the second gradient vector is equal to or greater than the reference value according to Equation 1, the switching unit 300 determines that the degree of change is less than the reference value.

[0090] As a result of the judgment in step S140, the degree of change is equal to the reference value. less than (The similarity is the reference value End ), the switching unit 300 proceeds to step S150 to delete the current training data set, and repeats steps S110 to S140 described above.

[0091] In contrast, as a result of the judgment in step S140, the degree of change is equal to or exceeds the reference value. End (The similarity is the reference value less than ), determine an update to the buffer training data set and proceed to step S160.

[0092] In step S160, the training unit 200 trains a gradient prediction model (GPM). The gradient prediction model (GPM) collects derived training data from training data used to train a current training model, and predicts gradient vectors of a future training model using the collected derived training data. Training of the gradient prediction model (GPM) will be described in more detail below.

[0093] Next, in step S170, the update unit 400 calculates the degree of influence of each of the plurality of learning data sets in the current learning data set on the prediction performance of the future learning model. of When a future learning model, which is a learning model, is trained using a current learning dataset, the influence of each of the multiple learning data in the current learning dataset on the predictive performance of the future learning model is calculated. The procedure for calculating such influences will be described in more detail below.

[0094] Next, in step S180, the update unit 400 selects a portion of the multiple learning data sets in the current learning data set in order of influence, and updates the buffer learning data set. At this time, the update unit 400 updates the buffer learning data set by deleting the buffer learning data previously stored in the buffer BF and storing the learning data selected from the multiple learning data in order of influence in the buffer BF. Also, in step S180, the remaining learning data not selected from the multiple learning data in order of influence is deleted.

[0095] By updating the buffer learning data set in the manner described above and continuously learning the learning model using the new learning data set together with the buffer learning data set, it is possible to overcome catastrophic forgetting while minimizing storage space and continuously provide a learning model that provides high accuracy at minimal cost.

[0096] Next, a method for training a gradient prediction model (GPM) according to an embodiment of the present invention will be described. Fig. 5 is a flowchart for explaining a method for training a gradient prediction model according to an embodiment of the present invention. Specifically, Fig. 5 is a detailed description of step S160 in Fig. 4.

[0097] 5, in step S210, the learning unit 200 selects a target layer of the current learning model. For example, it is assumed that the target layer is a convolutional layer. In step S210, the learning unit 200 divides a plurality of learning data of the current learning data set into epoch units and learns them while training the current learning model in step S120. Each time the learning unit 200 divides the learning data into epoch units and learns them, it stores a modified weight vector of the target layer of the current learning model.

[0098] As a result, in step S220, the learning unit 200 generates a plurality of derived learning data sets that are learning data derived from the current learning data set.

[0099] The derived learning data includes a weight vector of the target layer of the current learning model that is modified each time the current learning data set is divided into epoch units and the current learning model is trained iteratively, and a target gradient vector that indicates the difference between the weight vector from the previous epoch and the weight vector from the current epoch each time the current learning data set is divided into epoch units and the current learning model is trained iteratively.

[0100] Each time a plurality of learning data is divided into epoch units and learning is performed repeatedly, the weight vector of the target layer to be corrected changes as shown in the following Equation 2. [Number 2]

number

[0101] Here, n is the epoch index, LR denotes the learning rate, and G(n) denotes the gradient vector. In the present invention, the product of the learning rate LR and the gradient vector G(n) is approximated as the gradient vector. As a result, the weight vector may be expressed as [W(1), W(2), W(3), ..., W(E-1)] (E is an arbitrary positive constant). The target gradient vector corresponding to [W(1), W(2), W(3), ..., W(E-1)] may be expressed as [W(2)-W(1), W(3)-W(2), W(4)-W(3), ..., W(E)-W(E-1)].

[0102] Next, the learning unit 200 inputs the weight vectors of the derived learning data to a gradient prediction model (GPM) having untrained weights in step S230. Then, the gradient prediction model (GPM) performs a weight calculation in step S240, applying the untrained weights to the learning radiation image, and derives a weight difference vector that predicts the difference between the weight vector of the target layer in the current epoch and the weight vector in the previous epoch.

[0103] Next, in step S250, the learning unit 200 calculates a loss indicating the difference between the target gradient vector and the weighted difference vector using a loss function. Then, in step S260, the learning unit 200 performs optimization to correct the weights of the gradient prediction model (GPM) so that the loss derived by the loss function is minimized.

[0104] Thereafter, in step S270, the learning unit 200 determines whether a learning completion condition is satisfied. According to one embodiment, the learning completion condition may be when the loss calculated earlier (S250) converges and becomes equal to or less than a preset target value. If the learning completion condition is not satisfied as a result of the determination in step S270, the above-described steps S230 to S270 are repeated using multiple different training data. On the other hand, if the learning completion condition is satisfied as a result of the determination in step S270, the learning unit 200 completes learning of the gradient prediction model (GPM) in step S280.

[0105] Next, a method for calculating the degree of influence of each of multiple learning data sets in a current learning data set on the predictive performance of a future learning model according to an embodiment of the present invention will be described. Figure 6 is a flowchart for explaining a method for calculating the degree of influence of learning data sets on the predictive performance of a future learning model according to an embodiment of the present invention. That is, Figure 6 is a detailed description of step S170 in Figure 4.

[0106] In step S310, the update unit 400 calculates a plasticity score indicating the probability that the predicted value of the current learning model differs from the predicted value of the future learning model for the learning data of the current learning data set.

[0107] At this time, the update unit 400 can calculate the plasticity score according to the following Equation 3. [Number 3]

number

[0108] where:

number

number

number

number

[0109] On the other hand, since the future learning model has not yet been generated, the predicted value of the future learning model is estimated using the predicted gradient of the future learning model using a gradient prediction model (GP). As a result, the predicted value of the future learning model in Equation 3 is estimated by the following Equation 4. [Number 4]

number

[0110] where:

number

number

[0111] Next, in step S320, the update unit 400 compares the predicted value of the current learning model with the learning data of the current learning data set, and the value of the previous learning cycle of the current learning model. in The first past study session inA stability score is calculated that indicates the probability that the predicted value of the learned first past learning model differs from the predicted value.

[0112] At this time, the update unit 400 can calculate the stability score using the following Equation 5. [Number 5]

number

[0113] where:

number

[0114] n is the index of the training data in the current training data set, and t is the training cycle in the successive training. of is the index, and t is the current learning round. of t-1 is the first past learning session. of show.

number

number

number

[0115] Next, in step S330, the update unit 400 calculates the weighted average of the plasticity score and the stability score as the influence.

[0116] At this time, the update unit 400 can calculate the influence degree by the following Equation 6. [Number 6]

number

[0117] Here, S indicates the influence, i is the index of the training data in the current training dataset, λ is a preset weight, Plasticity means the plasticity score, and Stability means the stability score.

[0118] Next, a method for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention will be described. Figure 7 is a flowchart for explaining a method for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention. Figure 8 is an example of a screen for explaining a method for diagnosing defects using a continuous learning-based learning model according to an embodiment of the present invention.

[0119] In FIG. 7, the training model is the current training model trained using the previously stored buffer training data set and the current training data set as described above (S120).

[0120] 7, in step S410, the radiographic apparatus RTA performs radiography on an object having an image quality indicator (IQI) attached thereto, and outputs a radiographic inspection film capturing the image of the object having the IQI attached thereto. Here, the object may be a pipe, a tube, or the like.

[0121] Then, in step S420, the scanner SC scans the radiographic inspection film to generate a digitized radiographic image.

[0122] Once the digitized radiation image is generated in this manner, the generated radiation image is input to the inspection device 10. At this time, the radiation image may be input directly from the scanner SC, or may be stored in another storage medium and then input from the storage medium.

[0123] In step S430, the detection unit 500 of the inspection apparatus 10 receives a radiation image. Then, in step S440, the detection unit 500 detects defects in the radiation image using the latest model, i.e., the current learning model, among the learning models trained by continuous learning, as described above. At this time, the current learning model can detect the area occupied by the defect in the radiation image using a bounding box BB, as shown in FIG. 8.

[0124] Next, in step S450, the detection unit 500 outputs a report including the radiation image in which defects are detected using the bounding box BB.

[0125] 9 is a diagram illustrating a computing device according to an embodiment of the present invention. The computing device TN100 of FIG. 9 may be a device described herein (e.g., inspection device 10, storage device 40, etc.).

[0126] 9, computing device TN100 may include at least one processor TN110, a transceiver TN120, and a memory TN130. Computing device TN100 may further include a storage device TN140, an input interface device TN150, an output interface device TN160, etc. The components included in computing device TN100 may be connected by a bus TN170 to communicate with each other.

[0127] Processor TN110 can execute program commands stored in at least one of memory TN130 and storage device TN140. Processor TN110 can refer to a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. Processor TN110 can be configured to implement procedures, functions, methods, etc. described in connection with embodiments of the present invention. Processor TN110 can control each component of computing device TN100.

[0128] Memory TN130 and storage device TN140 can each store various information related to the operation of processor TN110. Memory TN130 and storage device TN140 can each be configured with at least one of a volatile storage medium and a non-volatile storage medium. For example, memory TN130 may be configured with at least one of a read-only memory (ROM) and a random access memory (RAM).

[0129] The transceiver TN120 can transmit or receive wired or wireless signals and can communicate with a network.

[0130] In particular, the collection unit 100, learning unit 200, switching unit 300, update unit 400, and detection unit 500 of the inspection device 10 according to the embodiment of the present invention may be realized in the form of a program readable by a computer means, stored in memory TN130, and executed by processor TN110. Alternatively, the collection unit 100, learning unit 200, switching unit 300, update unit 400, and detection unit 500 may be lower modules of processor TN110.

[0131] Furthermore, the buffer BF according to the embodiment of the present invention may be the memory TN130 or storage device TN140 of the inspection device 10 or storage device 40.

[0132] Meanwhile, the various methods according to the above-described embodiments of the present invention can be realized in the form of a program readable by various computer means and recorded on a computer-readable recording medium. Here, the recording medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the recording medium may be specially designed and constructed for the present invention, or may be known and available to those skilled in the art of computer software. For example, recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include not only machine language, such as that produced by a compiler, but also high-level language codes executable by a computer using an interpreter, etc. Such hardware devices can be configured to operate as one or more software modules to perform the operations of the present invention, or vice versa.

[0133] Although one embodiment of the present invention has been described above, a person having ordinary skill in the art may modify and change the present invention in various ways by adding, changing, deleting or adding components within the scope of the concept of the present invention as set forth in the claims, and this also falls within the scope of the present invention.

Claims

1. a learning unit learning a current learning model using a buffer learning data set and a current learning data set stored in advance in a buffer; a switching unit calculating a degree of change from a plurality of past learning models in learning times two or more before a current learning model in the continuous learning; a step of the switching unit determining whether the degree of change is equal to or greater than a reference value; the switching unit determining an update to the buffer training data set if the degree of change is equal to or greater than the reference value; a detection unit detecting defects in a radiation image using the current learning model when the radiation image is input; Including, A method for diagnosing defects.

2. A method for diagnosing defects as described in claim 1, wherein the step of calculating the degree of change includes a step in which the switching unit derives the degree of change based on the similarity between two or more gradient vectors, each of which indicates a change between each weight vector corresponding to the current learning model and the multiple past learning models.

3. If the similarity is less than the reference value, the degree of change is equal to or greater than the reference value; If the similarity is equal to or greater than the reference value, the change rate is less than the reference value; 3. The method for diagnosing defects of claim 2, further comprising the step of: said switching unit erasing said current training data set if said degree of change is less than said reference value.

4. The step of training the current training model includes: the learning unit loads the buffer learning data set, which is learning data selected from past learning data sets in successive learning according to the degree of influence on the prediction performance of the current learning model; training the current training model using the buffer training data set and the current training data set; Including, 10. The method for diagnosing defects of claim 1.

5. When an update unit uses the current learning dataset to train a future learning model to be trained in a learning round next to the current learning model in continuous learning, the update unit calculates the degree of influence of each of the multiple learning data of the current learning dataset on the predictive performance of the future learning model; the update unit selecting a portion of the plurality of training data sets in accordance with the order of the influence levels and updating the buffer training data set; further comprising:

10. The method for diagnosing defects of claim 1.

6. The step of calculating the influence degree includes: The update unit calculates a plasticity score indicating a probability that a predicted value of the current learning model and a predicted value of the future learning model differ for learning data of the current learning data set; The update unit calculates a stability score indicating the probability that a predicted value of the current learning model differs from a predicted value of a first past learning model trained in a first past learning round, which is a learning round before the current learning model, for the learning data of the current learning data set; The update unit calculates a weighted average of the plasticity score and the stability score as the influence degree; Including, 6. A method for diagnosing defects according to claim 5.

7. The step of calculating the stability score includes: Formula [Number 73] is calculated by [Number 74] is the stability score, [Number 75] is the training data of the current training data set, [Number 76] is a predicted value of the first past learning model, [Number 77] is the predicted value of the current learning model, 7. A method for diagnosing defects according to claim 6.

8. The step of calculating the plasticity score includes: Formula [Number 78] is calculated by [Number 79] is the plasticity score, [Number 80] is the training data of the current training data set, [Number 81] is the predicted value of the future learning model, [Number 82] is the predicted value of the current learning model, 7. A method for diagnosing defects according to claim 6.

9. The step of calculating the plasticity score includes: Formula [Number 83] By Derive a predicted value of the future learning model; [Number 84] is the predicted value of the future learning model, x is training data of the current training data set, GP (θ ｔ ) is the gradient vector predicted by the future learning model through the gradient prediction model; 9. A method for diagnosing defects according to claim 8.

10. The step of calculating the influence degree includes: Formula [Number 85] Calculated by S is the influence level, i is the index of the current training data, λ is the weight, Plasticity is the plasticity score, Stability is the stability score.

6. A method for diagnosing defects according to claim 5.

11. The step of calculating the degree of change includes: The degree of change is derived according to the similarity between a first gradient vector indicating the change between the weight vector of the current learning model and the weight vector of a first past learning model learned in a learning session prior to the current learning model, and a second gradient vector indicating the change between the weight vector of the first past learning model and the weight vector of a second past learning model learned in a learning session prior to the first past learning model. A method for diagnosing defects according to any one of claims 1 to 10.

12. The similarity is Formula [Number 86] Calculated by CS is the similarity, g ｔ is the first gradient vector between the weight vector of the current training model and the weight vector of the first past training model, g ｔ－１ is the second gradient vector between the weight vector of the first past-trained model and the weight vector of the second past-trained model, g Ｔ ｔ－１ is the transpose vector of the second gradient vector, 12. A method for diagnosing defects according to claim 11.

13. a learning unit that learns a current learning model using a buffer learning data set pre-stored in a buffer and a current learning data set; calculating a degree of change from a plurality of past learning models in learning times two or more before the current learning model in the continuous learning; determining whether the degree of change is equal to or greater than a reference value; a switching unit that determines an update to the buffer learning data set if the degree of change is equal to or greater than the reference value; a detection unit that, when a radiation image is input, detects defects in the radiation image using the current learning model; Including, A device for diagnosing defects.

14. The learning unit loading the buffer training data set, which is training data selected from past training data sets in accordance with the degree of influence on the prediction performance of the current training model in continuous training; training the current training model using the buffer training data set and the current training data set; 14. An apparatus for diagnosing defects according to claim 13.

15. When a future learning model to be learned in a learning round following the current learning model in continuous learning is trained using the current learning data set, the degree of influence of each of the multiple learning data sets of the current learning data set on the predictive performance of the future learning model is calculated, an update unit that selects a portion of the plurality of training data sets in accordance with the order of the influence levels and updates the buffer training data set; further comprising:

14. An apparatus for diagnosing defects according to claim 13.

16. The update unit Calculating a plasticity score indicating the probability that the predicted value of the current learning model differs from the predicted value of the future learning model for the learning data of the current learning data set; calculating a stability score indicating the probability that the predicted value of the current learning model differs from the predicted value of a first past learning model trained in a first past learning round that is the learning round before the current learning model, for the learning data of the current learning data set; A weighted average of the plasticity score and the stability score is calculated as the influence degree.

16. An apparatus for diagnosing defects according to claim 15.

17. The update unit Formula [Number 87] The stability score is calculated by [Number 88] is the stability score, [Number 89] is the training data of the current training data set, [Number 90] is a predicted value of the first past learning model, [Number 91] is the predicted value of the current learning model, 17. An apparatus for diagnosing defects according to claim 16.

18. The update unit Formula [Number 92] The plasticity score is calculated by [Number 93] is the plasticity score, [Number 94] is the training data of the current training data set, [Number 95] is the predicted value of the future learning model, [Number 96] is the predicted value of the current learning model, 17. An apparatus for diagnosing defects according to claim 16.

19. The update unit Formula [Number 97] By Derive a predicted value of the future learning model; [Number 98] is the predicted value of the future learning model, x is training data of the current training data set, GP (θ ｔ ) is the gradient vector predicted by the future learning model through the gradient prediction model; 20. An apparatus for diagnosing defects according to claim 18.

20. The update unit Formula [Number 99] The influence degree is calculated by S is the influence level, i is the index of the current training data, λ is the weight, Plasticity is the plasticity score, Stability is the stability score.

16. An apparatus for diagnosing defects according to claim 15.

21. The switching unit The degree of change is derived according to the similarity between a first gradient vector indicating the change between the weight vector of the current learning model and the weight vector of a first past learning model learned in a learning session prior to the current learning model, and a second gradient vector indicating the change between the weight vector of the first past learning model and the weight vector of a second past learning model learned in a learning session prior to the first past learning model.

21. An apparatus for diagnosing defects according to any one of claims 15 to 20.

22. The switching unit Formula [Number 100] The similarity is calculated by CS is the similarity, g ｔ is the first gradient vector between the weight vector of the current training model and the weight vector of the first past training model, g ｔ－１ is the second gradient vector between the weight vector of the first past-trained model and the weight vector of the second past-trained model, g Ｔ ｔ－１ is the transpose vector of the second gradient vector, 22. An apparatus for diagnosing defects according to claim 21.

Images