KR102393951B1 - Object-oriented data augmentation method - Google Patents

Abstract

An object-centric data augmentation method performed by a computing device is disclosed. The object-oriented data augmentation method includes: identifying a first region of training data; determining a first maximum value and a first minimum value among brightness values of pixels included in the first area; determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value; wherein the second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value with an equal function, and the second maximum value is the second minimum value and the first maximum value.

Description

OBJECT-ORIENTED DATA AUGMENTATION METHOD

The present invention relates to a data augmentation method, and more particularly to a data augmentation method for improving the object recognition rate in a TOF camera.

In deep neural networks, the accuracy of the model can be increased by increasing the number of parameters by stacking many hidden layers. To properly train with up to millions of parameters, a huge amount of training data is required. In particular, the learning data must be diverse enough to reflect reality well enough, and the quality must be excellent.

If there is not enough training data to train the parameters, overfitting problems that hinder the performance of the model are highly likely to occur. Overfitting refers to a phenomenon in which the model over-adapts only to the training data and fails to respond properly to the test data or new data. In order to solve this overfitting problem, it is most important to train the network with a variety of learning data.

As such, data augmentation is used as one of the techniques to secure the data necessary to sufficiently train the deep neural network. Data augmentation refers to a methodology for acquiring a large amount of new training data by artificially changing a small amount of training data. For example, new image data may be obtained by flipping or cropping the image.

A method of changing the brightness of an image is also one of the data augmentation methodologies. In general, data augmentation is performed by adjusting the brightness of the entire image.

In the case of a TOF camera image, in general, an object to be detected is bright because it is close to it, whereas the background is very dark because it is relatively far away. According to the characteristics of the TOF camera image, when the data augmentation method for collectively adjusting the brightness of the entire image is used, the change in brightness of the target to be detected is inevitably small. Accordingly, the effect of data augmentation may be insignificant, and the recognition rate improvement effect may also be small.

Therefore, a data augmentation method suitable for TOF camera images is required.

The present disclosure has been devised in response to the above background art, and is to provide an object-oriented data augmentation method suitable for a TOF camera image.

An object-oriented data augmentation method performed by a computing device, the method comprising: identifying a first region of training data; determining a first maximum value and a first minimum value among brightness values of pixels included in the first area; determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value; wherein the second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value with an equal function, and the second maximum value is the second minimum value and the first maximum value.

In addition, the determining of the second maximum value and the second minimum value for data augmentation using the first maximum value and the first minimum value includes: a random function using the first maximum value and the first minimum value determining the second maximum value and the second minimum value through may include

In addition, the step of determining the second maximum value and the second minimum value for data augmentation using the first maximum value and the first minimum value includes: the second maximum value and the second minimum value according to the following equation (1) The method may further include determining a second minimum value.

Equation (1)

X ₁ : first minimum value

X ₂ : first maximum value

New X ₁ : 2nd minimum value

New X ₂ : 2nd maximum value

M: maximum allowed

U: Uniform random number

Round(): rounding function

In addition, adjusting the brightness value of all pixels included in the learning data; may further include.

The adjusting of the brightness of all pixels included in the training data may include: adjusting the brightness of all pixels by a difference between the first maximum value and the second maximum value; may include

The method may further include: assigning a maximum or minimum allowable value to a pixel that is out of an allowable brightness value among pixels included in the training data; may further include.

Also, the training data may be an 8-bit image, the maximum allowable value may be 255, and the allowable minimum value may be 0.

Also, the first area may be an object to be detected in the image.

Also, the first region may be a face part.

In addition, the learning data may include at least one of an RGB camera image, a ToF camera image, and an NIR photographed image.

Disclosed is a computer program stored in a computer-readable storage medium for solving the above problems. The computer program includes instructions for causing one or more processors to perform an object-centric data augmentation method, the instructions comprising: identifying a first region of training data; determining a first maximum value and a first minimum value among brightness values of pixels included in the first area; determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value; wherein the second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value with an equal function, and the second maximum value is the second minimum value and the first maximum value.

Disclosed is a computing device for performing an object-oriented data augmentation method for solving the above-described problems. The computing device may include: a memory; and a processor; wherein the processor is configured to: identify a first region of training data; determining a first maximum value and a first minimum value among brightness values of pixels included in the first area; determine a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value, the second minimum value being the maximum allowed value of the first maximum value and the first minimum value is determined by processing a value obtained by subtracting the difference as an equal function, wherein the second maximum value is determined based on the sum of the second minimum value and the first maximum value; The brightness values of pixels included in the first area may be re-determined according to the second maximum value and the second minimum value.

Disclosed is a computer-readable recording medium storing a data structure corresponding to a parameter of a neural network that is at least partially updated in a learning process for solving the above-described problem. The operation of the neural network is based at least in part on the parameters, and the learning process uses at least training data on which an object-oriented data augmentation method has been performed, the method comprising: identifying a first region of the training data; determining a first maximum value and a first minimum value among brightness values of pixels included in the first area; determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value; wherein the second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value with an equal function, and the second maximum value is the second minimum value and the first maximum value.

The present disclosure may provide an object-centric data augmentation method.

1 is a block diagram of a computing device for a processor to perform an object-oriented data augmentation method according to some embodiments of the present disclosure;
2 is a schematic diagram illustrating a network function according to some embodiments of the present disclosure;
3 is a flowchart illustrating an object-oriented data augmentation method according to some embodiments of the present disclosure.
4A to 4C are diagrams for explaining an object-oriented data augmentation method according to some embodiments of the present disclosure in comparison with an existing method.
5A to 5C illustrate training data images according to each step of an object-oriented data augmentation method according to some embodiments of the present disclosure.
6 is experimental data for explaining a comparison between an existing data augmentation method and an object-oriented data augmentation method according to some embodiments of the present disclosure.
7 is a simplified, general schematic diagram of an exemplary computing environment in which embodiments of the present disclosure may be implemented.

Various embodiments are now described with reference to the drawings. In this specification, various descriptions are presented to provide an understanding of the present disclosure. However, it is apparent that these embodiments may be practiced without these specific descriptions.

The terms “component,” “module,” “system,” and the like, as used herein, refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or execution of software. For example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device may be a component. One or more components may reside within a processor and/or thread of execution. A component may be localized within one computer. A component may be distributed between two or more computers. In addition, these components can execute from various computer readable media having various data structures stored therein. Components may communicate via a network such as the Internet with another system, for example, via a signal having one or more data packets (eg, data and/or signals from one component interacting with another component in a local system, distributed system, etc.) may communicate via local and/or remote processes depending on the data being transmitted).

In addition, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from context, "X employs A or B" is intended to mean one of the natural implicit substitutions. That is, X employs A; X employs B; or when X employs both A and B, "X employs A or B" may apply to either of these cases. It should also be understood that the term “and/or” as used herein refers to and includes all possible combinations of one or more of the listed related items.

Also, the terms "comprises" and/or "comprising" should be understood to mean that the feature and/or element in question is present. However, it should be understood that the terms "comprises" and/or "comprising" do not exclude the presence or addition of one or more other features, elements and/or groups thereof. Also, unless otherwise specified or unless it is clear from context to refer to a singular form, the singular in the specification and claims should generally be construed to mean “one or more”.

And, the term "at least one of A or B" should be interpreted as meaning "when including only A", "when including only B", and "when combined with the configuration of A and B".

Those skilled in the art will further appreciate that the various illustrative logical blocks, configurations, modules, circuits, means, logics, and algorithm steps described in connection with the embodiments disclosed herein may be implemented in electronic hardware, computer software, or combinations of both. It should be recognized that they can be implemented with To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, configurations, means, logics, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application. However, such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

Descriptions of the presented embodiments are provided to enable those skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art of the present disclosure. The generic principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Therefore, the present invention is not limited to the embodiments presented herein. This invention is to be interpreted in its widest scope consistent with the principles and novel features presented herein.

In the present disclosure, a network function and an artificial neural network and a neural network may be used interchangeably.

1 is a block diagram of a computing device for an object-oriented data augmentation method according to an embodiment of the present disclosure.

The configuration of the computing device 100 shown in FIG. 1 is only a simplified example. In an embodiment of the present disclosure, the computing device 100 may include other components for performing the computing environment of the computing device 100 , and only some of the disclosed components may configure the computing device 100 .

The computing device 100 may include a processor 110 , a memory 120 , and a network unit (not shown).

The processor 110 may include one or more cores, and a central processing unit (CPU) of a computing device, a general purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU). unit) and the like, and may include a processor for data analysis and deep learning. The processor 110 may read a computer program stored in a memory and perform data processing for machine learning according to an embodiment of the present disclosure. According to an embodiment of the present disclosure, the processor 110 may perform an operation for learning the neural network. The processor 110 for learning of the neural network, such as processing input data for learning in deep learning (DL), extracting features from the input data, calculating an error, updating the weight of the neural network using backpropagation calculations can be performed. At least one of a CPU, a GPGPU, and a TPU of the processor 110 may process learning of a network function. For example, the CPU and the GPGPU can process learning of a network function and data classification using the network function. Also, in an embodiment of the present disclosure, learning of a network function and data classification using the network function may be processed by using the processors of a plurality of computing devices together. In addition, the computer program executed in the computing device according to an embodiment of the present disclosure may be a CPU, GPGPU or TPU executable program.

According to an embodiment of the present disclosure, the memory 120 may store any type of information generated or determined by the processor 110 and any type of information received by the network unit.

According to an embodiment of the present disclosure, the memory 120 is a flash memory type, a hard disk type, a multimedia card micro type, and a card type memory (eg, a SD or XD memory, etc.), Random Access Memory (RAM), Static Random Access Memory (SRAM), Read-Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Programmable Read (PROM) -Only Memory), a magnetic memory, a magnetic disk, and an optical disk may include at least one type of storage medium. The computing device 100 may operate in relation to a web storage that performs a storage function of the memory 120 on the Internet. The description of the above-described memory is only an example, and the present disclosure is not limited thereto.

Network unit according to an embodiment of the present disclosure: Public Switched Telephone Network (PSTN), x Digital Subscriber Line (xDSL), Rate Adaptive DSL (RADSL), Multi Rate DSL (MDSL), Very High Speed DSL (VDSL) ), a variety of wired communication systems such as Universal Asymmetric DSL (UADSL), High Bit Rate DSL (HDSL), and Local Area Network (LAN) can be used.

In addition, the network unit presented in the present specification (Code Division Multi Access), TDMA (Time Division Multi Access), FDMA (Frequency Division Multi Access), OFDMA (Orthogonal Frequency Division Multi Access), SC-FDMA (Single Carrier-FDMA) ) and other systems may be used.

In the present disclosure, the network unit may be configured regardless of its communication mode, such as wired and wireless, and may be configured with various communication networks, such as a personal area network (PAN) and a wide area network (WAN). In addition, the network may be a well-known World Wide Web (WWW), and may use a wireless transmission technology used for short-range communication, such as infrared (IrDA) or Bluetooth (Bluetooth).

The techniques described herein may be used in the networks mentioned above, as well as in other networks.

2 is a schematic diagram illustrating a network function according to an embodiment of the present disclosure;

Throughout this specification, computational model, neural network, network function, and neural network may be used interchangeably. A neural network may be composed of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network is configured to include at least one or more nodes. Nodes (or neurons) constituting the neural networks may be interconnected by one or more links.

In the neural network, one or more nodes connected through a link may relatively form a relationship between an input node and an output node. The concepts of an input node and an output node are relative, and any node in an output node relationship with respect to one node may be in an input node relationship in a relationship with another node, and vice versa. As described above, an input node-to-output node relationship may be created around a link. One or more output nodes may be connected to one input node through a link, and vice versa.

In the relationship between the input node and the output node connected through one link, the value of the data of the output node may be determined based on data input to the input node. Here, a link interconnecting the input node and the output node may have a weight. The weight may be variable, and may be changed by the user or algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are interconnected to one output node by respective links, the output node sets values input to input nodes connected to the output node and links corresponding to the respective input nodes. An output node value may be determined based on the weight.

As described above, in a neural network, one or more nodes are interconnected through one or more links to form an input node and an output node relationship in the neural network. The characteristics of the neural network may be determined according to the number of nodes and links in the neural network, the correlation between the nodes and the links, and the value of a weight assigned to each of the links. For example, when the same number of nodes and links exist and there are two neural networks having different weight values of the links, the two neural networks may be recognized as different from each other.

A neural network may consist of a set of one or more nodes. A subset of nodes constituting the neural network may constitute a layer. Some of the nodes constituting the neural network may configure one layer based on distances from the initial input node. For example, a set of nodes having a distance of n from the initial input node may constitute n layers. The distance from the initial input node may be defined by the minimum number of links that must be passed to reach the corresponding node from the initial input node. However, the definition of such a layer is arbitrary for description, and the order of the layer in the neural network may be defined in a different way from the above. For example, a layer of nodes may be defined by a distance from the final output node.

The initial input node may mean one or more nodes to which data is directly input without going through a link in a relationship with other nodes among nodes in the neural network. Alternatively, in a relationship between nodes based on a link in a neural network, it may mean nodes that do not have other input nodes connected by a link. Similarly, the final output node may refer to one or more nodes that do not have an output node in relation to other nodes among nodes in the neural network. In addition, the hidden node may mean nodes constituting the neural network other than the first input node and the last output node.

The neural network according to an embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be the same as the number of nodes in the output layer, and the number of nodes decreases and then increases again as progresses from the input layer to the hidden layer. can Also, in the neural network according to another embodiment of the present disclosure, the number of nodes in the input layer may be less than the number of nodes in the output layer, and the number of nodes may be reduced as the number of nodes progresses from the input layer to the hidden layer. there is. In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes in the input layer may be greater than the number of nodes in the output layer, and the number of nodes increases as the number of nodes progresses from the input layer to the hidden layer. can The neural network according to another embodiment of the present disclosure may be a neural network in a combined form of the aforementioned neural networks.

A deep neural network (DNN) may refer to a neural network including a plurality of hidden layers in addition to an input layer and an output layer. Deep neural networks can be used to identify the latent structures of data. In other words, it can identify the potential structure of photos, texts, videos, voices, and music (e.g., what objects are in the photos, what the text and emotions are, what the texts and emotions are, etc.) . Deep neural networks include convolutional neural networks (CNNs), recurrent neural networks (RNNs), auto encoders, generative adversarial networks (GANs), and restricted boltzmann machines (RBMs). machine), a deep belief network (DBN), a Q network, a U network, a Siamese network, and a Generative Adversarial Network (GAN). The description of the deep neural network described above is only an example, and the present disclosure is not limited thereto.

In an embodiment of the present disclosure, the network function may include an autoencoder. The auto-encoder may be a kind of artificial neural network for outputting output data similar to input data. The auto encoder may include at least one hidden layer, and an odd number of hidden layers may be disposed between the input/output layers. The number of nodes in each layer may be reduced from the number of nodes of the input layer to an intermediate layer called the bottleneck layer (encoding), and then expanded symmetrically with reduction from the bottleneck layer to the output layer (symmetrically with the input layer). The auto-encoder can perform non-linear dimensionality reduction. The number of input layers and output layers may correspond to a dimension after preprocessing the input data. In the auto-encoder structure, the number of nodes of the hidden layer included in the encoder may have a structure that decreases as the distance from the input layer increases. If the number of nodes in the bottleneck layer (the layer with the fewest nodes located between the encoder and decoder) is too small, a sufficient amount of information may not be conveyed, so a certain number or more (e.g., more than half of the input layer, etc.) ) may be maintained.

The neural network may be trained using at least one of supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning. Learning of the neural network may be a process of applying knowledge for the neural network to perform a specific operation to the neural network.

A neural network can be trained in a way that minimizes output errors. In the training of a neural network, iteratively input the training data into the neural network, calculate the output of the neural network and the target error for the training data, and calculate the error of the neural network from the output layer of the neural network to the input layer in the direction to reduce the error. It is a process of updating the weight of each node in the neural network by backpropagation in the direction. In the case of teacher learning, learning data in which the correct answer is labeled in each learning data is used (ie, labeled learning data), and in the case of comparative learning, the correct answer may not be labeled in each learning data. That is, for example, learning data in the case of teacher learning related to data classification may be data in which categories are labeled in each of the learning data. Labeled training data is input to the neural network, and an error can be calculated by comparing the output (category) of the neural network with the label of the training data. As another example, in the case of comparison learning related to data classification, an error may be calculated by comparing the input training data with the neural network output. The calculated error is back propagated in the reverse direction (ie, from the output layer to the input layer) in the neural network, and the connection weight of each node of each layer of the neural network may be updated according to the back propagation. The change amount of the connection weight of each node to be updated may be determined according to a learning rate. The computation of the neural network on the input data and the backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently according to the number of repetitions of the learning cycle of the neural network. For example, in the early stage of learning of a neural network, a high learning rate can be used to enable the neural network to quickly obtain a certain level of performance, thereby increasing efficiency, and using a low learning rate at a later stage of learning can increase accuracy.

In the training of neural networks, in general, the training data may be a subset of real data (that is, data to be processed using the trained neural network), and thus, the error on the training data is reduced, but the error on the real data is reduced. There may be increasing learning cycles. Overfitting is a phenomenon in which errors on actual data increase by over-learning on training data as described above. For example, a phenomenon in which a neural network that has learned a cat by seeing a yellow cat does not recognize that it is a cat when it sees a cat other than yellow may be a type of overfitting. Overfitting can act as a cause of increasing errors in machine learning algorithms. In order to prevent such overfitting, various optimization methods can be used. In order to prevent overfitting, methods such as increasing the training data, regularization, and dropout that deactivate some of the nodes of the network in the process of learning, and the use of a batch normalization layer are applied. can

A computer-readable medium storing a data structure is disclosed according to an embodiment of the present disclosure.

The data structure may refer to the organization, management, and storage of data that enables efficient access and modification of data. A data structure may refer to an organization of data to solve a specific problem (eg, data retrieval, data storage, and data modification in the shortest time). A data structure may be defined as a physical or logical relationship between data elements designed to support a particular data processing function. The logical relationship between data elements may include a connection relationship between user-defined data elements. Physical relationships between data elements may include actual relationships between data elements physically stored on a computer-readable storage medium (eg, persistent storage). A data structure may specifically include a set of data, relationships between data, and functions or instructions applicable to data. Through an effectively designed data structure, the computing device can perform an operation while using the resource of the computing device to a minimum. Specifically, the computing device may increase the efficiency of operations, reads, insertions, deletions, comparisons, exchanges, and retrievals through effectively designed data structures.

A data structure may be classified into a linear data structure and a non-linear data structure according to the type of the data structure. The linear data structure may be a structure in which only one piece of data is connected after one piece of data. The linear data structure may include a list, a stack, a queue, and a deck. A list may mean a set of data in which an order exists internally. The list may include a linked list. The linked list may be a data structure in which data is linked in such a way that each data is linked in a line with a pointer. In a linked list, a pointer may contain information about a link with the next or previous data. A linked list may be expressed as a single linked list, a doubly linked list, or a circularly linked list according to a shape. A stack can be a data enumeration structure with limited access to data. A stack can be a linear data structure in which data can be manipulated (eg, inserted or deleted) at only one end of the data structure. The data stored in the stack may be a data structure (LIFO-Last in First Out). A queue is a data listing structure that allows limited access to data, and unlike a stack, it may be a data structure that is stored later (FIFO-First in First Out). A deck can be a data structure that can process data at either end of the data structure.

The non-linear data structure may be a structure in which a plurality of data is connected after one data. The nonlinear data structure may include a graph data structure. A graph data structure may be defined as a vertex and an edge, and the edge may include a line connecting two different vertices. A graph data structure may include a tree data structure. The tree data structure may be a data structure in which one path connects two different vertices among a plurality of vertices included in the tree. That is, it may be a data structure that does not form a loop in the graph data structure.

Throughout this specification, computational model, neural network, network function, and neural network may be used interchangeably. Hereinafter, the neural network is unified and described. The data structure may include a neural network. And the data structure including the neural network may be stored in a computer-readable medium. Data structures, including neural networks, also include preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation functions associated with each node or layer of the neural network, and neural networks. It may include a loss function for learning of . A data structure comprising a neural network may include any of the components disclosed above. That is, the data structure including the neural network includes preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation function associated with each node or layer of the neural network, and neural network It may be configured to include all or any combination thereof, such as a loss function for learning of . In addition to the above-described configurations, a data structure including a neural network may include any other information that determines a characteristic of a neural network. In addition, the data structure may include all types of data used or generated in the operation process of the neural network, and is not limited to the above. Computer-readable media may include computer-readable recording media and/or computer-readable transmission media. A neural network may be composed of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network is configured to include at least one or more nodes.

The data structure may include data input to the neural network. A data structure including data input to the neural network may be stored in a computer-readable medium. The data input to the neural network may include learning data input in a neural network learning process and/or input data input to the neural network in which learning is completed. Data input to the neural network may include pre-processing data and/or pre-processing target data. The preprocessing may include a data processing process for inputting data into the neural network. Accordingly, the data structure may include data to be pre-processed and data generated by pre-processing. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

The data structure may include the weights of the neural network. (In this specification, weight and parameter may be used interchangeably.) And the data structure including the weight of the neural network may be stored in a computer-readable medium. The neural network may include a plurality of weights. The weight may be variable, and may be changed by the user or algorithm in order for the neural network to perform a desired function. For example, when one or more input nodes are interconnected to one output node by respective links, the output node sets values input to input nodes connected to the output node and links corresponding to the respective input nodes. A data value output from the output node may be determined based on the weight. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

By way of example and not limitation, the weight may include a weight variable in a neural network learning process and/or a weight in which neural network learning is completed. The variable weight in the neural network learning process may include a weight at a time point at which a learning cycle starts and/or a weight variable during the learning cycle. The weight for which neural network learning is completed may include a weight for which a learning cycle is completed. Accordingly, the data structure including the weights of the neural network may include a data structure including the weights that vary in the neural network learning process and/or the weights on which the neural network learning is completed. Therefore, it is assumed that the above-described weights and/or combinations of weights are included in the data structure including the weights of the neural network. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

The data structure including the weights of the neural network may be stored in a computer-readable storage medium (eg, memory, hard disk) after being serialized. Serialization can be the process of converting a data structure into a form that can be reconstructed and used later by storing it on the same or a different computing device. The computing device may serialize the data structure to send and receive data over the network. A data structure including weights of the serialized neural network may be reconstructed in the same computing device or in another computing device through deserialization. The data structure including the weight of the neural network is not limited to serialization. Furthermore, the data structure including the weights of the neural network is a data structure to increase computational efficiency while using the resources of the computing device to a minimum (e.g., B-Tree, Trie, m-way search tree, AVL tree, Red-Black Tree). The foregoing is merely an example, and the present disclosure is not limited thereto.

The data structure may include hyper-parameters of the neural network. In addition, the data structure including the hyperparameters of the neural network may be stored in a computer-readable medium. The hyper parameter may be a variable variable by a user. Hyperparameters are, for example, learning rate, cost function, number of iterations of the learning cycle, weight initialization (e.g., setting the range of weight values to be initialized for weights), Hidden Unit The number (eg, the number of hidden layers, the number of nodes of the hidden layer) may be included. The above-described data structure is merely an example, and the present disclosure is not limited thereto.

3 is a flowchart illustrating an object-oriented data augmentation method according to an embodiment of the present disclosure.

According to the object-oriented data augmentation method of the present disclosure, the object recognition rate of the object recognition model can be increased by adjusting the brightness centering on the object to be identified, rather than collectively adjusting the overall brightness of the image, which is the learning data. In particular, in the case of the ToF camera image, the background is very dark compared to the object. Accordingly, in the case of collectively adjusting the brightness of the entire image, the effect of data augmentation may be insignificant because the brightness change of the object to be identified is insignificant. According to the object-oriented data augmentation method of the present disclosure, since the brightness of the object to be identified can be adjusted in various ranges, the effect of data augmentation is increased, and thus the object recognition rate effect of the neural network model learned with the learning data is also increased. can be

According to some embodiments of the present disclosure, an object-oriented data augmentation method may include steps s100 to s150 described below. However, these steps are described for the purpose of illustration only, and some steps may be changed or omitted, or additional steps may be added. Also, these steps may be performed in any order and are not necessarily performed in the order described below.

According to some embodiments of the present disclosure, the object-oriented data augmentation method may include identifying a first region from the learning data (s100). The first area may be an area (or object) to be detected in the image that is the training data. For example, the first region may be an object to be detected in the image. The first region may be plural when there are plural objects present on the image. Here, the first area may be displayed as a boundary of an object to be detected. For example, the first region may be displayed as a boundary of an object within a bounding box. Recognizing the first region may mean recognizing a labeled region (or object) on the training data.

In one embodiment, the first region may be a face part in an image for training a model for detecting a face part (facial boundary). In another embodiment, the first region may be a lesion on an image (eg, an X-ray image) to train a model for detecting the lesion. The second area may include an area on the learning data except for the first area. However, the present invention is not limited thereto, and the first region and the second region may be defined in various ways according to the purpose of the neural network.

An object detection method using a neural network can be classified into a technique based on region extraction (RoI pooling, region of interest pooling) and a technique based on a grid space (Grid Cell). Among them, a technique based on region extraction may be a method of estimating the size and location of a bounding box to detect a specific object in one feature map. Since the bounding box contains the boundary of the object, the estimated bounding box can be used to detect the object.

In order to increase the accuracy of determining the first region (or bounding box) of the neural network model, data augmentation may be performed in such a way that the brightness value of the first region is changed to more various values. To this end, the object-oriented data augmentation method of the present disclosure may randomly adjust the pixel value of the training data based on the brightness range (ie, the first maximum value and the first minimum value) of the first region.

In some embodiments of the present disclosure, the object-oriented data augmentation method of the present disclosure may include determining a first maximum value and a first minimum value among brightness values of pixels included in the first area ( S110 ). Here, the first maximum value may mean a maximum value among brightness values of pixels included in the first area before data augmentation is performed. Specifically, for example, the first maximum value may be determined as the highest value among the brightness values of pixels included in the first area. As another example, the first maximum value may be determined as one of the top n values of the brightness values of pixels included in the first area (eg, the second value among the top 5). As another example, either a predetermined number of pixels (eg, five) do not have a particular upper value, or a predetermined percentage of pixels (eg, pixels corresponding to 1% of the first area) do not have a particular upper value. In the case of not having , the second-order value may be determined as the first maximum value. However, the present invention is not limited thereto, and the first maximum value may be determined by various methods for indicating the brightness range of the first area.

Also, the first minimum value may mean a minimum value among brightness values of pixels included in the first area before data augmentation is performed. Specifically, for example, the first minimum value may be determined as the lowest value among the brightness values of pixels included in the first area. As another example, the first minimum value may be determined as one of the lower n values of the brightness values of the pixels (eg, a third value among the ten lower values). As another example, when a predetermined number of pixels (eg, 7) do not have a specific sub-value or a predetermined percentage of pixels (eg, 5% of the first area) do not have a specific sub-value. In this case, the difference lower value of the specific lower value may be determined as the first minimum value. However, the present invention is not limited thereto, and the first minimum value may be determined by various methods for indicating the brightness range of the first area.

Although the object-oriented data augmentation method according to the present disclosure is described with respect to a brightness value (ie, brightness), the object-oriented data augmentation method according to the present disclosure may be applied to other factors (eg, color or saturation).

By determining the first maximum value and the first minimum value among the brightness values of the pixels included in the first area, the range of the brightness values of the pixels included in the first area may be determined. The first maximum and first minimum values (that is, the brightness values of pixels included in the first region without considering the second region) so that the brightness values of the pixels included in the first region have various values according to data augmentation. range) can be used to augment data. By performing data augmentation by identifying the range of the brightness value of the first area, it is possible to prevent the change of the brightness value of the first area to be detected from being restricted by the second area (eg, a relatively dark background). there is.

In some embodiments of the present disclosure, the object-oriented data augmentation method of the present disclosure may include determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value ( s120). Here, the second maximum value may mean a maximum value among brightness values of pixels included in the first area after data augmentation is performed. Also, the second minimum value may mean a minimum value among brightness values of pixels included in the first area before data augmentation is performed. The second maximum value and the second minimum value may have various values so that data augmentation is performed.

In some embodiments of the present disclosure, the step of determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value comprises using the first maximum value and the first minimum value to generate a random function. and determining a second maximum value and a second minimum value through Data augmentation with respect to the training data may be performed by assigning a random value. Specifically, by randomly determining the second maximum value and the second minimum value through a random function, various learning data may be generated. The random function may be implemented in various ways known in the art.

The second maximum value and the second minimum value generated through the random function may be determined within a certain range according to the type of image that is the training data. For example, when the training data is an 8-bit image having a brightness value of 0 to 255, the second maximum value is determined to have a value of 255-X ₂ +X ₁ to 255, and the second minimum value is 0 to X ₁ . Here, X ₁ may be a first minimum value, and X ₂ may mean a first maximum value. However, the present invention is not limited thereto, and the second maximum value and the second maximum value may have various ranges.

According to some embodiments of the present disclosure, the determining of the second maximum value and the second minimum value using the first maximum value and the first minimum value includes determining the second maximum value and the second minimum value according to Equation 1 below. may include the step of

X ₁ : first minimum value

X ₂ : first maximum value

New X ₁ : 2nd minimum value

New X ₂ : 2nd maximum value

M: maximum allowed

U: Uniform random number

Round(): rounding function

Here, the uniform random number may have any value between 0 and 1. M may mean a maximum brightness value allowed in an image that is training data. For example, when the image, which is the training data, is an 8-bit image, the brightness value may have a value of 0 to 255. Accordingly, when the image, which is the training data, is an 8-bit image, the maximum allowable value M may be 255. In this equation, the minimum allowable value is defined as 0. 'Round()' may be a rounding function. For example, 'Round()' can be used to take an integer value. However, the present invention is not limited thereto, and the second maximum value and the second minimum value may be randomly determined according to the first maximum value and the first minimum value in various ways.

According to some embodiments of the present disclosure, the method for augmenting object-oriented data may include re-determining brightness values of pixels included in the first area according to a second maximum value and a second minimum value ( S130 ).

Specifically, brightness values of pixels included in the first area may be re-determined according to newly determined maximum and minimum values (ie, second maximum and second minimum values) to perform data augmentation. For example, when the second maximum value is determined to be 240 and the second minimum value is determined to be 120, the brightness values of pixels included in the first area between the first maximum value and the first minimum value are between 240 and 120. It can be recrystallized in proportion to the value. Since the brightness value determined in this way has a different value from the original training data, a data augmentation effect can be provided. However, the present invention is not limited thereto, and brightness values of pixels included in the first region may be re-determined in various ways.

According to some embodiments of the present disclosure, the object-oriented data augmentation method of the present disclosure may include adjusting the brightness values of all pixels included in the training data (s140). For example, the brightness value of all pixels included in the training data may be adjusted to a random value (eg, an arbitrary value between -20% and 20%). By randomly adjusting the brightness values of all pixels, data augmentation of the entire image may be achieved. However, the present invention is not limited thereto, and the brightness values of all pixels included in the training data may be adjusted in various ways. Here, all pixels generally mean pixels included in the first area and the second area, but according to some embodiments, all pixels may mean pixels included only in the second area except for the first area.

According to some embodiments of the present disclosure, adjusting the brightness of all pixels included in the learning data may include adjusting the brightness of all pixels by a difference between the first maximum value and the second maximum value. .

Specifically, the brightness of the entire image (eg, the brightness value of all pixels) may be adjusted by the difference between the first maximum value and the second maximum value. For example, when the first maximum value is 200 and the second maximum value is 210, the difference between the first maximum value and the second maximum value may be -10. Accordingly, the brightness value of all pixels may be adjusted by -10. In this case, the case in which pixels included in the first area have brightness values in an unacceptable range can be reduced.

In general, a ToF camera image or NIR photographed image (but not limited thereto, an image photographed with a general camera (eg, an RGB camera image) may have a characteristic that the object is bright at close range, whereas the background is very dark. Therefore, the background is very dark and generally consists of pixels having a low brightness value, and the object to be detected is often composed of pixels having a relatively high brightness value compared to the background. It may be desirable to adjust the brightness mainly according to the amount of change in the maximum value of interest, but the present invention is not limited thereto, and the brightness value of all pixels may be adjusted in various ways.

According to some embodiments of the present disclosure, the object-oriented data augmentation method of the present disclosure may include giving a maximum allowable value or an allowable minimum value to a pixel that deviates from an allowable brightness value among pixels included in the training data. can (s150).

For example, after recrystallization of the above-described brightness value, the brightness values of pixels included in the training data that are 8-bit images may have values that are not allowed in the 8-bit image (ie, values outside the range of 0 to 255). In this case, the maximum allowable value is collectively given to pixels having a brightness value exceeding the allowable maximum value (that is, the brightness value 255), and the maximum allowable value (ie, the brightness value 0) is less than A minimum allowable value may be collectively assigned to pixels having a brightness value. As another example, when the training data is a 16-bit image, the maximum allowed value may be 65535, and the minimum value may be 0. In this case, the maximum allowable value is collectively given to pixels having a brightness value exceeding the allowable maximum value (that is, the brightness value 65535), and the brightness that is less than the allowable minimum value (that is, the brightness value 0) A minimum allowable value may be collectively assigned to pixels having a value. However, the present invention is not limited thereto, and brightness values of pixels outside the allowable range may be adjusted in various ways according to the type of learning data.

According to some embodiments of the present disclosure, the training data may include at least one of an RGB camera image, a ToF camera image, and an NIR photographed image. In general, a TOF camera image or an NIR photographed image is bright because an object to be detected is at a close distance, whereas the background has a very dark feature because it is relatively far away. The object-oriented data augmentation method of the present disclosure identifies a brightness range of an object to be detected (ie, a first maximum value and a first minimum value of a first region) and performs data augmentation so that the object has brightness values in various ranges. Therefore, it may be particularly advantageous when used for ToF camera images or NIR-captured images. However, since an object captured by a general camera (eg, an RGB camera image) also has an object that is brighter than the background in many cases, the object-oriented data augmentation method of the present disclosure can provide an improved effect even for a general image. The object-oriented data augmentation method of the present disclosure is not limited thereto and may be used for various types of images.

The object-oriented data augmentation method of the present disclosure may be used together with various existing data augmentation methods. For example, the object-oriented data augmentation method of the present disclosure may be used in combination with existing methods such as left-right inversion, vertical inversion, position movement, rotation, size adjustment, and cropping of only a portion of a large image of an image.

4A to 4C are diagrams for explaining the object-oriented data augmentation method of the present disclosure in comparison with the existing method.

Figure 4a shows the original image before data augmentation. For example, the image of FIG. 4A may be used as training data for training a neural network for determining the bounding box 200 to detect a human face.

FIG. 4B shows an image in which a conventional data augmentation method is applied to the image of FIG. 4A. The image shown in FIG. 4B is an image in which the overall brightness of the image of FIG. 4A is collectively increased (20% of the total brightness). According to the existing method, when the overall brightness is collectively increased, the brightness of the object to be detected becomes too high, so that the effect of data augmentation on the learning data may be lowered. In addition, if the overall brightness is lowered at once, the background may become too dark, and the effect of data augmentation of the training data may be lowered.

4C illustrates an image to which the object-oriented data augmentation method of the present disclosure is applied to the image of FIG. 4A . It can be seen that the brightness change for the face of the person shown in the image of FIG. 4C is large compared to the background. Since the object-oriented data augmentation method of the present disclosure adjusts the brightness value of the entire image after first adjusting the brightness value of the object in consideration of the brightness value of the object, the object (ie, the first area) and the background (ie, the second region) may have more combinations of brightness values compared to the existing method. Therefore, according to the object-oriented data augmentation method of the present disclosure, since the brightness of the object to be identified can be adjusted in various ranges, the effect of data augmentation can be increased. Accordingly, the object recognition rate effect of the neural network model trained with such training data may also be increased.

5A to 5C illustrate training data images according to each step of an object-oriented data augmentation method according to some embodiments of the present disclosure.

According to some embodiments of the present disclosure, the object-oriented data augmentation method of the present disclosure may be applied to training data for training a neural network for determining the bounding box 200 to detect a human face.

FIG. 5A shows the first region 210 identified according to the step s100 of identifying the first region from the training data. In the object-oriented data augmentation method of the present disclosure, the brightness of the first area is first adjusted using the brightness range of the first area (that is, the first maximum value and the first minimum value among the brightness values of pixels included in the first area). Afterwards, the brightness of the second area may be adjusted.

5B shows a step of determining a first maximum value and a first minimum value among the brightness values of pixels included in the first area (s110), a second maximum value for data augmentation using the first maximum value and the first minimum value, and An image is shown in which the step (s120) of determining the second minimum value and the step (s130) of re-determining the brightness values of pixels included in the first area according to the second maximum value and the second minimum value are performed. Referring to FIG. 5B , it can be seen that although the brightness of the first area 210 is adjusted, the brightness of the second area 220 is not. Accordingly, in the object-oriented data augmentation method of the present disclosure, the brightness value of the first area may be determined in various ranges by first adjusting the brightness value of the first area without considering the second area.

5C shows the step of adjusting the brightness values of all pixels included in the training data (s140) and the steps of giving a maximum value or a minimum allowable value to a pixel that is out of an allowable brightness value among pixels included in the training data An image in which (s150) is performed is shown. Data augmentation of the entire image may be achieved by first adjusting the brightness values of the first area and then adjusting the brightness values of all pixels including the second area.

According to the object-oriented data augmentation method of the present disclosure, since the brightness of the object to be identified can be adjusted in various ranges, the effect of data augmentation is increased, and thus the object recognition rate effect of the neural network model learned with the learning data is also increased. can be

6 is experimental data for explaining the comparison between the existing data augmentation method and the object-oriented data augmentation method according to an embodiment of the present disclosure.

The training data used in the experiment were 925 black-and-white images (Width x Height = 320x240) captured by the ToF camera and 3,500 face images (Width x Height = 640x480) captured by NIR.

The training data was labeled by marking 67 face landmarks using a face landmark labeling tool (GUI). The size of the input layer is set to Width x Height x Channel = 256 x 256 x 1.

In the case of data augmentation technique, 7 techniques were used, and 6 techniques were used equally in both cases. However, in the case of image brightness change, the existing technique used a method of collectively changing the entire image between -20% and +20%, whereas the object-oriented data augmentation method according to the present disclosure is the method described with reference to FIG. 3 above. was used.

The training method consisted of Batch size: 64, Momentum: 0.9, Decay rate: 0.0005, Learning Rate: 0.001, epoch: 100, Dropout Rate: 0.2, and data augmentation was performed 10 times the original data.

As for the training results, the final loss values were 0.72 (train), 0.69 (test), recall = 0.86, and Precision = 0.97 in the case of the existing technique. In contrast, in the case of the object-oriented data augmentation method according to the present disclosure, the final loss values are 0.76 (train), 0.72 (test), recall = 0.95, Precision = 0.98, and the object-oriented data augmentation method according to the present disclosure is the existing technique It can be seen that the results are improved compared to .

7 is a simplified, general schematic diagram of an exemplary computing environment in which embodiments of the present disclosure may be implemented.

Although the present disclosure has been described above as being generally capable of being implemented by a computing device, those skilled in the art will appreciate that the present disclosure may be implemented in hardware and software in combination with computer-executable instructions and/or other program modules and/or in combination with computer-executable instructions and/or other program modules that may be executed on one or more computers. It will be appreciated that it can be implemented as a combination.

Generally, program modules include routines, programs, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In addition, those skilled in the art will appreciate that the methods of the present disclosure can be applied to single-processor or multiprocessor computer systems, minicomputers, mainframe computers as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, and the like. It will be appreciated that each of these may be implemented in other computer system configurations, including those capable of operating in connection with one or more associated devices.

The described embodiments of the present disclosure may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

Computers typically include a variety of computer-readable media. Any medium accessible by a computer can be a computer-readable medium, and such computer-readable media includes volatile and nonvolatile media, transitory and non-transitory media, removable and non-transitory media. including removable media. By way of example, and not limitation, computer-readable media may include computer-readable storage media and computer-readable transmission media. Computer readable storage media includes volatile and nonvolatile media, temporary and non-transitory media, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. includes media. A computer-readable storage medium may be RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage device, magnetic cassette, magnetic tape, magnetic disk storage device, or other magnetic storage device. device, or any other medium that can be accessed by a computer and used to store the desired information.

Computer-readable transmission media typically embodies computer-readable instructions, data structures, program modules or other data, etc. in a modulated data signal such as a carrier wave or other transport mechanism, and Includes any information delivery medium. The term modulated data signal means a signal in which one or more of the characteristics of the signal is set or changed so as to encode information in the signal. By way of example, and not limitation, computer-readable transmission media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer-readable transmission media.

An example environment 1100 implementing various aspects of the disclosure is shown including a computer 1102 , the computer 1102 including a processing unit 1104 , a system memory 1106 , and a system bus 1108 . do. A system bus 1108 couples system components, including but not limited to system memory 1106 , to the processing device 1104 . The processing device 1104 may be any of a variety of commercially available processors. Dual processor and other multiprocessor architectures may also be used as processing unit 1104 .

The system bus 1108 may be any of several types of bus structures that may further interconnect a memory bus, a peripheral bus, and a local bus using any of a variety of commercial bus architectures. System memory 1106 includes read only memory (ROM) 1110 and random access memory (RAM) 1112 . A basic input/output system (BIOS) is stored in non-volatile memory 1110, such as ROM, EPROM, EEPROM, etc., the BIOS is the basic input/output system (BIOS) that helps transfer information between components within computer 1102, such as during startup. contains routines. RAM 1112 may also include high-speed RAM, such as static RAM, for caching data.

The computer 1102 may also include an internal hard disk drive (HDD) 1114 (eg, EIDE, SATA) - this internal hard disk drive 1114 may also be configured for external use within a suitable chassis (not shown). Yes—a magnetic floppy disk drive (FDD) 1116 (eg, for reading from or writing to removable diskette 1118), and an optical disk drive 1120 (eg, a CD-ROM) for reading from, or writing to, disk 1122, or other high capacity optical media, such as DVD. The hard disk drive 1114 , the magnetic disk drive 1116 , and the optical disk drive 1120 are connected to the system bus 1108 by the hard disk drive interface 1124 , the magnetic disk drive interface 1126 , and the optical drive interface 1128 , respectively. ) can be connected to The interface 1124 for implementing an external drive includes at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

These drives and their associated computer-readable media provide non-volatile storage of data, data structures, computer-executable instructions, and the like. In the case of computer 1102, drives and media correspond to storing any data in a suitable digital format. Although the description of computer readable media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art will use zip drives, magnetic cassettes, flash memory cards, cartridges, etc. It will be appreciated that other tangible computer-readable media such as etc. may also be used in the exemplary operating environment and any such media may include computer-executable instructions for performing the methods of the present disclosure.

A number of program modules may be stored in the drive and RAM 1112 , including an operating system 1130 , one or more application programs 1132 , other program modules 1134 , and program data 1136 . All or portions of the operating system, applications, modules, and/or data may also be cached in RAM 1112 . It will be appreciated that the present disclosure may be implemented in various commercially available operating systems or combinations of operating systems.

A user may enter commands and information into the computer 1102 via one or more wired/wireless input devices, for example, a pointing device such as a keyboard 1138 and a mouse 1140 . Other input devices (not shown) may include a microphone, IR remote control, joystick, game pad, stylus pen, touch screen, and the like. Although these and other input devices are connected to the processing unit 1104 through an input device interface 1142 that is often connected to the system bus 1108, parallel ports, IEEE 1394 serial ports, game ports, USB ports, IR interfaces, It may be connected by other interfaces, etc.

A monitor 1144 or other type of display device is also coupled to the system bus 1108 via an interface, such as a video adapter 1146 . In addition to the monitor 1144, the computer typically includes other peripheral output devices (not shown), such as speakers, printers, and the like.

Computer 1102 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) 1148 via wired and/or wireless communications. Remote computer(s) 1148 may be workstations, computing device computers, routers, personal computers, portable computers, microprocessor-based entertainment devices, peer devices, or other common network nodes, and are typically connected to computer 1102 . Although it includes many or all of the components described for it, only memory storage device 1150 is shown for simplicity. The logical connections shown include wired/wireless connections to a local area network (LAN) 1152 and/or a larger network, eg, a wide area network (WAN) 1154 . Such LAN and WAN networking environments are common in offices and companies, and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a worldwide computer network, for example, the Internet.

When used in a LAN networking environment, the computer 1102 is coupled to the local network 1152 through a wired and/or wireless communication network interface or adapter 1156 . Adapter 1156 may facilitate wired or wireless communication to LAN 1152 , which LAN 1152 also includes a wireless access point installed therein for communicating with wireless adapter 1156 . When used in a WAN networking environment, the computer 1102 may include a modem 1158, be connected to a communication computing device on the WAN 1154, or establish communications over the WAN 1154, such as over the Internet. have other means. A modem 1158 , which may be internal or external and a wired or wireless device, is coupled to the system bus 1108 via a serial port interface 1142 . In a networked environment, program modules described for computer 1102 , or portions thereof, may be stored in remote memory/storage device 1150 . It will be appreciated that the network connections shown are exemplary and other means of establishing a communication link between the computers may be used.

Computer 1102 may be associated with any wireless device or object that is deployed and operates in wireless communication, for example, a printer, scanner, desktop and/or portable computer, portable data assistant (PDA), communication satellite, wireless detectable tag. It operates to communicate with any device or place and phone. This includes at least Wi-Fi and Bluetooth wireless technologies. Accordingly, the communication may be a predefined structure as in a conventional network or may simply be an ad hoc communication between at least two devices.

Wi-Fi (Wireless Fidelity) makes it possible to connect to the Internet, etc. without a wired connection. Wi-Fi is a wireless technology such as cell phones that allows these devices, eg, computers, to transmit and receive data indoors and outdoors, ie anywhere within range of a base station. Wi-Fi networks use a radio technology called IEEE 802.11 (a, b, g, others) to provide secure, reliable, and high-speed wireless connections. Wi-Fi can be used to connect computers to each other, to the Internet, and to wired networks (using IEEE 802.3 or Ethernet). Wi-Fi networks may operate in unlicensed 2.4 and 5 GHz radio bands, for example at 11 Mbps (802.11a) or 54 Mbps (802.11b) data rates, or in products that include both bands (dual band). .

One of ordinary skill in the art of this disclosure will understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, instructions, information, signals, bits, symbols, and chips that may be referenced in the above description are voltages, currents, electromagnetic waves, magnetic fields or particles, optical field particles or particles, or any combination thereof.

Those of ordinary skill in the art of the present disclosure will recognize that the various illustrative logical blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein include electronic hardware, (convenience For this purpose, it will be understood that it may be implemented by various forms of program or design code (referred to herein as software) or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. A person skilled in the art of the present disclosure may implement the described functionality in various ways for each specific application, but such implementation decisions should not be interpreted as a departure from the scope of the present disclosure.

The various embodiments presented herein may be implemented as methods, apparatus, or articles of manufacture using standard programming and/or engineering techniques. The term article of manufacture includes a computer program, carrier, or media accessible from any computer-readable storage device. For example, computer-readable storage media include magnetic storage devices (eg, hard disks, floppy disks, magnetic strips, etc.), optical disks (eg, CDs, DVDs, etc.), smart cards, and flash drives. memory devices (eg, EEPROMs, cards, sticks, key drives, etc.). Also, various storage media presented herein include one or more devices and/or other machine-readable media for storing information.

It is understood that the specific order or hierarchy of steps in the presented processes is an example of exemplary approaches. Based on design priorities, it is to be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of the present disclosure. The appended method claims present elements of the various steps in a sample order, but are not meant to be limited to the specific order or hierarchy presented.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments presented herein, but is to be construed in the widest scope consistent with the principles and novel features presented herein.

Claims (13)

An object-oriented data augmentation method performed by a computing device, comprising:
identifying a first region of the training data;
determining a first maximum value and a first minimum value among brightness values of pixels included in the first area;
determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and
recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value;
including,
The second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value as an equal function;
the second maximum value is determined based on a sum of the second minimum value and the first maximum value;
An object-centric data augmentation method.
The method of claim 1,
Determining the second maximum value and the second minimum value for data augmentation using the first maximum value and the first minimum value may include:
determining the second maximum value and the second minimum value through a random function using the first maximum value and the first minimum value;
containing,
An object-centric data augmentation method.
The method of claim 1,
Determining the second maximum value and the second minimum value for data augmentation using the first maximum value and the first minimum value may include:
determining the second maximum value and the second minimum value according to the following equation (1):
further comprising,
An object-centric data augmentation method.
Equation (1)

X ₁ : first minimum value
X ₂ : first maximum value
New X ₁ : 2nd minimum value
New X ₂ : 2nd maximum value
M: maximum allowed
U: Uniform random number
Round(): rounding function
The method of claim 1,
adjusting brightness values of all pixels included in the training data;
further comprising,
An object-centric data augmentation method.
5. The method of claim 4,
The step of adjusting the brightness of all pixels included in the training data includes:
adjusting the brightness values of all the pixels by the difference between the first maximum value and the second maximum value;
containing,
An object-centric data augmentation method.
The method of claim 1,
assigning an allowable maximum value or an allowable minimum value to a pixel that is out of an allowable brightness value among pixels included in the training data;
further comprising,
An object-centric data augmentation method.
7. The method of claim 6,
the training data is an 8-bit image, the maximum allowed value is 255, and the minimum allowed value is 0;
An object-centric data augmentation method.
The method of claim 1,
The first area is an object to be detected in the image,
An object-centric data augmentation method.
9. The method of claim 8,
The first region is the facial region,
An object-centric data augmentation method.
The method of claim 1,
The learning data includes at least one of an RGB camera image, a ToF camera image, or an NIR photographed image,
An object-centric data augmentation method.
A computer program stored on a computer-readable storage medium, the computer program comprising instructions for causing one or more processors to perform an object-centric data augmentation method, the instructions comprising:
identifying a first region of the training data;
determining a first maximum value and a first minimum value among brightness values of pixels included in the first area;
determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and
recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value;
including,
The second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value with an equal function;
the second maximum value is determined based on a sum of the second minimum value and the first maximum value;
A computer program stored in a computer-readable storage medium.
A computing device for performing an object-centric data augmentation method, the computing device comprising:
Memory; and
processor;
including,
The processor is:
identify a first region of training data;
determining a first maximum value and a first minimum value among brightness values of pixels included in the first area;
determine a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value, the second minimum value being the maximum allowed value of the first maximum value and the first minimum value is determined by processing a value obtained by subtracting the difference as an equal function, wherein the second maximum value is determined based on the sum of the second minimum value and the first maximum value;
recrystallizing the brightness values of pixels included in the first area according to the second maximum value and the second minimum value;
A computing device that performs object-centric data augmentation.
A computer-readable recording medium storing a data structure corresponding to a parameter of a neural network that is at least partially updated in a learning process, wherein the operation of the neural network is based at least in part on the parameter, and the learning process is an object-oriented data augmentation method Using at least this performed training data, the method comprising:
identifying a first region of the training data;
determining a first maximum value and a first minimum value among brightness values of pixels included in the first area;
determining a second maximum value and a second minimum value for data augmentation using the first maximum value and the first minimum value; and
recrystallizing brightness values of pixels included in the first area according to the second maximum value and the second minimum value;
including,
The second minimum value is determined by processing a value obtained by subtracting a difference between the first maximum value and the first minimum value from an allowable maximum value as an equal function;
the second maximum value is determined based on a sum of the second minimum value and the first maximum value;
A computer-readable recording medium having a data structure stored thereon.

Images