Electronic device and data selection method thereof

Abstract

An electronic device and a data selection method thereof are provided. The electronic device includes a camera that obtains an image and a processor that detects an object from the image. The processor estimates a distance of the detected object, determines a weight based on the estimated distance, applies the determined weight to determine an entropy, and determines whether to obtain data of the detected object based on the determined entropy.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Korean Patent Application No. 10-2023-0156606, filed on November 13, 2023, which is hereby incorporated by reference in its entirety for all purposes. Technical Field

[0003] The present disclosure relates to an electronic device and a data selection method thereof. Background Art

[0004] Active learning is the task of selecting data that may be most effective for training a model on multiple pieces of unlabeled data. Entropy-based active learning is widely used, but only to address the problem of imbalanced data per class. In other words, entropy-based active learning is used to select images of objects belonging to classes for which there is missing object data.

[0005] When obtaining ground truth (GT) for detecting three-dimensional (3D) objects when there are objects near the camera, many pieces of short-range object data are obtained because distant objects are covered by the objects. Due to the provided configuration, an imbalance problem between short-range data and long-range data and imbalance of data for each category occur.

[0006] The information included in this background of the disclosure is only for enhancement of understanding of the general background of the disclosure and should not be taken as an acknowledgment or any form of suggestion that this information forms the prior art already known to a person skilled in the art. Summary of the invention

[0007] Various aspects of the present disclosure are directed to providing an electronic device for obtaining object data per distance and object data per category in a balanced manner and a data selection method thereof.

[0008] The technical problems to be solved by the present disclosure are not limited to the above-mentioned problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art to which the present disclosure belongs from the following description.

[0009] According to one aspect of the present disclosure, an electronic device may include: a camera that obtains an image; and a processor that is configured to detect an object from the image. The processor may estimate a distance of the detected object, may be configured to determine a weight based on the estimated distance, may determine entropy by using the determined weight, and may be configured to determine whether to obtain data of the detected object based on the determined entropy.

[0010] The processor may refer to the lookup table, search for a frame size that is most similar to the size of the detected object, and may estimate a distance mapped to the found frame size as the distance of the detected object.

[0011] The processor may match and classify objects having a size most similar to a previously obtained object for each predetermined frame size, may be configured to determine an average distance of the classified objects for each frame size, and may be configured to determine the determined average distance as the object distance according to the frame size.

[0012] The electronic device may further include a memory operatively connected to the processor and storing a lookup table in which object distances of the frame size are defined.

[0013] The processor may set a data acquisition target rate corresponding to the estimated distance, may verify a data acquisition current rate corresponding to the estimated distance, and may be configured to determine a weight using the data acquisition target rate and the data acquisition current rate.

[0014] The processor is further configured to determine an entropy by using the determined weights and a probability value of a probability that the detected object will belong to a predetermined category.

[0015] The processor is configured to determine whether the entropy is greater than or equal to a predetermined reference value, and may select the data of the detected object as data required to generate ground truth (GT) data based on a conclusion that the entropy is greater than or equal to the reference value.

[0016] The processor can use a deep learning network to detect objects from images.

[0017] The deep learning network may output a probability value for the probability that the detected object will belong to the i-th category.

[0018] According to another aspect of the present disclosure, a data selection method of an electronic device may include: obtaining an image using a camera; detecting an object from the image; estimating a distance of the detected object; determining a weight based on the estimated distance; determining entropy by using the determined weight; and determining whether to obtain data of the detected object based on the determined entropy.

[0019] Estimation of the distance of the object may include referring to a lookup table, searching for a box size that is most similar to the size of the detected object, and estimating the distance mapped to the found box size as the distance of the detected object.

[0020] The data selection method may further include: matching and classifying objects having a size most similar to previously obtained objects for each predetermined frame size; determining an average distance of the classified objects for each frame size; and determining the determined average distance as the object distance according to the frame size.

[0021] The data selection method may further include generating the lookup table using the object distance according to the frame size and storing the lookup table in a memory operably connected to the processor.

[0022] Determining the weight may include setting a data acquisition target rate corresponding to the estimated distance, verifying a data acquisition current rate corresponding to the estimated distance, and determining the weight using the data acquisition target rate and the data acquisition current rate.

[0023] Determining the entropy includes determining the entropy by using the determined weights and a probability value of a probability that the detected object will belong to a predetermined category.

[0024] Determining whether to obtain data of the object may include determining whether entropy is greater than or equal to a predetermined reference value, and based on a conclusion that the entropy is greater than or equal to the reference value, selecting the detected data of the object as data required to generate ground truth (GT) data.

[0025] Detecting data may include detecting objects from images using a deep learning network.

[0026] The detection data may include a probability value output by the deep learning network that the detected object will belong to the i-th category.

[0027] The methods and apparatus of the present disclosure have other features and advantages that will be apparent from or set forth in more detail in the accompanying drawings and the following detailed description incorporated herein, which together serve to explain certain principles of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a block diagram illustrating a configuration of an electronic device according to various exemplary embodiments of the present disclosure;

[0029] Figure 2 is a diagram describing object distance determination according to frame size according to various exemplary embodiments of the present disclosure;

[0030] Figure 3 is a flowchart illustrating a data selection method of an electronic device according to various exemplary embodiments of the present disclosure;

[0031] Figure 4A is a view showing an example of a detection object according to various exemplary embodiments of the present disclosure;

[0032] Figure 4B is a diagram describing a distance map for each object according to various exemplary embodiments of the present disclosure;

[0033] Figure 4C is a diagram describing weight determination according to various exemplary embodiments of the present disclosure;

[0034] Figure 5 is a diagram describing the effects of a data selection method according to various exemplary embodiments of the present disclosure; and

[0035] Figure 6 is a block diagram illustrating a computing system for executing a data selection method according to various exemplary embodiments of the present disclosure.

[0036] It will be appreciated that the drawings are not necessarily drawn to scale, and that they present somewhat simplified representations of various features illustrating the basic principles of the present disclosure. The specific design features of the present disclosure as included herein (including, for example, specific dimensions, orientations, positions, and shapes) and shapes will be determined in part by the specific intended application and use environment.

[0037] In the drawings, reference numbers refer to the same or equivalent parts of the present disclosure throughout the several figures of the drawing. DETAILED DESCRIPTION

[0038] Reference will now be made in detail to various embodiments of the present disclosure, examples of which are shown in the accompanying drawings and described below. Although the present disclosure will be described in conjunction with exemplary embodiments of the present disclosure, it should be understood that this specification is not intended to limit the present disclosure to those exemplary embodiments of the present disclosure. On the other hand, the present disclosure is intended to cover not only the exemplary embodiments of the present disclosure, but also various replacements, modifications, equivalents and other embodiments that may be included in the spirit and scope of the present disclosure as defined by the appended claims.

[0039] Hereinafter, various exemplary embodiments of the present disclosure will be described in detail with reference to the exemplary drawings. When adding reference numerals to the components of the drawings, it should be noted that in the case where the same constituent elements are shown in other drawings, the components are also represented by the same reference numerals. In addition, detailed descriptions of well-known features or functions will be excluded so as not to unnecessarily obscure the subject matter of the present disclosure.

[0040] When describing the components of the exemplary embodiments of the present disclosure, the terms first, second, A, B, (a), (b), etc. may be used herein. These terms are only used to distinguish one component from another component, without limiting the corresponding components, and are not related to the order or priority of the corresponding components. In addition, unless otherwise defined, all terms including technical and scientific terms used herein include the same meanings as commonly understood by those skilled in the art to which the present disclosure belongs. Those terms as defined in commonly used dictionaries should be interpreted as having the same meaning as the contextual meaning in the relevant field, and should not be interpreted as having an ideal or overly formal meaning unless explicitly defined as having an ideal or overly formal meaning in this application.

[0041] Figure 1 is a block diagram illustrating a configuration of an electronic device according to various exemplary embodiments of the present disclosure.

[0042] The electronic device 100 may be installed in a vehicle. Figure 1 As shown, the electronic device 100 may include a camera 110 , a user interface 120 , a memory 130 , and a processor 140 .

[0043] The camera 110 may obtain an image of its surroundings. For example, when the camera 110 is mounted on a vehicle, it may capture the surroundings of the vehicle. The camera 110 may store the captured image in the memory 130, or may directly send the captured image to the processor 140.

[0044] The camera 110 may include at least one of an image sensor, such as a charge coupled device (CCD) image sensor, a complementary metal oxide semiconductor (CMOS) image sensor, a charge injection device (CPD) image sensor, or a charge injection device (CID) image sensor. The camera 110 may include an image processor for performing image processing (such as noise removal, file compression, image quality adjustment, and / or saturation adjustment) on an image obtained by the image sensor.

[0045] The user interface 120 may be a device that helps the electronic device 100 and the user interact with each other. The user interface 120 may include an input device (e.g., a keyboard, a touch pad, a microphone, a touch screen, etc.) for generating data according to the user's manipulation, an output device (e.g., a display, a speaker, a tactile signal output device, etc.) for outputting information according to the operation of the electronic device 100, and the like.

[0046] The memory 130 may store a plurality of pieces of frame size information predetermined by a system designer. The memory 130 may store a lookup table in which a distance according to a predetermined frame size is defined. The memory 130 may store a deep learning network, a data selection algorithm, etc. The memory 130 may include training data, ground truth (GT) data, etc. The GT data may be used as verification data. The memory 130 may store pre-set setting information.

[0047] The memory 130 may be a non-transitory storage medium storing instructions executed by the processor 140. The memory 130 may include at least one of storage media such as a flash memory, a hard disk, a solid state disk (SSD), a random access memory (RAM), a static RAM (SRAM), a read-only memory (ROM), a programmable ROM (PROM), an electrically erasable programmable ROM (EEPROM), or an erasable programmable ROM (EPROM).

[0048] The processor 140 may be configured to control the overall operation of the electronic device 100. The processor 140 may include at least one of a processing device such as an application specific integrated circuit (ASIC), a digital signal processor (DSP), a programmable logic device (PLD), a field programmable gate array (FPGA), a central processing unit (CPU), a microcontroller, or a microprocessor.

[0049] The processor 140 may use a previously acquired dataset of an object (e.g., a car, a bus, a truck, etc.) to statistically analyze the distance according to the frame size. The dataset may include multiple pieces of data of the object. The data of each piece of data (i.e., object data) may include the object size, the distance (or object distance) from a reference point (e.g., the position where the camera 110 is set, the position of the bumper of the vehicle, etc.), the object category, etc.

[0050] The processor 140 may compare the object size included in the previously obtained object data with the predetermined frame size. Here, the object size may include the width and height of the bounding box surrounding the object. The processor 140 may match the frame size with the object size having the most similar size. In other words, the processor 140 may classify the object data for each object size with reference to the predetermined frame size.

[0051] The processor 140 may be configured to determine an average distance of the object using the object distances included in the plurality of classified object data. In other words, the processor 140 may be configured to determine an average distance of the object matching the predetermined frame size. The processor 140 may define the object distance of each predetermined frame size using the determined average distance to generate a lookup table.

[0052] The processor 140 may obtain an image through the camera 110. The processor 140 may input the obtained image into a deep learning network (or a deep learning model). The deep learning network may detect an object from the image and may be configured to determine a probability value for each category for the detected object. For example, the deep learning network may be configured to determine and output each of the probability that the object detected from the image will be a vehicle, the probability that the object will be a bus, and the probability that the object will be a truck.

[0053] The processor 140 may estimate the distance (or target distance) from the reference point to the detected object based on the size of the object detected by the deep learning network. The processor 140 may refer to the lookup table stored in the memory 130, search for a box size similar to the size of the bounding box of the detected object, and may map the distance according to the found box size.

[0054] The processor 140 may set a target rate (or data acquisition target rate) at which data of an object corresponding to a mapping distance is expected to be acquired. The target rate may be preset by a system designer. The data acquisition target rate refers to the ratio of the target number of object images of a given size to the number of object images acquired for each of different sizes. The processor 140 may verify the data acquisition current rate (or current rate) of data of an object corresponding to a mapping distance. The data acquisition current rate refers to the ratio between the number of object images acquired for each size by class classification, entropy-based active learning (i.e., before applying entropy weights).

[0055] Processor 140 may be configured to use a target rate R tar and the current rate R to determine the adaptive weight W. The adaptive weight (or weight) W may be expressed as Equation 1 below.

[0056] [Equation 1]

[0057]

[0058] The processor 140 may apply the determined weight W to determine the entropy E of the detected object. The processor 140 may be configured to determine the entropy E using Equation 2 below.

[0059] [Equation 2]

[0060]

[0061] Here, P i can be defined as the probability that an object will belong to the i-th category.

[0062] The processor 140 may compare the determined entropy E with a predetermined reference value (or reference entropy), and may be configured to determine whether to select the object data based on the comparison result. The processor 140 may be configured to determine whether the determined entropy E is greater than or equal to the predetermined reference value. When it is determined that the determined entropy E is greater than or equal to the predetermined reference value, the processor 140 may select the data of the object corresponding to the entropy (or object data) as the data required to generate the GT data (or data to be labeled).

[0063] Thereafter, the processor 140 may be configured to generate GT data of the selected object. The processor 140 may use the generated GT data to train a three-dimensional (3D) object detection model stored in the memory 130. The processor 140 may store the trained 3D object detection model in the memory 130.

[0064] Figure 2 is a diagram for describing object distance determination according to frame size according to various exemplary embodiments of the present disclosure.

[0065] The electronic device 100 may statistically analyze the object distance according to the frame size using previously acquired pieces of data of an object (eg, a car, a bus, a truck, etc.).

[0066] First, the processor 140 of the electronic device 100 may store a plurality of predetermined frame size information 210 input from outside the electronic device 100 in the memory 130. The frame size information 210 may include a two-dimensional (2D) size of the frame, ie, a width and a height of the frame.

[0067] The processor 140 may compare the size of the previously obtained object with the predetermined frame size information 210. Here, the size of the object may include the width and height of the bounding box surrounding the object. The processor 140 may match the frame size that is most similar to the size of the object. In other words, the processor 140 may classify the objects matched for each predetermined frame size (refer to reference numeral 220).

[0068] The processor 140 may be configured to use the distances of the multiple classified object data to determine the average distance of the multiple classified objects. In other words, the processor 140 may be configured to determine the average distance of the objects matching the predetermined frame size. The processor 140 may use the determined average distance to define the object distance of each predetermined frame size to generate the lookup table 230.

[0069] Figure 3 is a flowchart illustrating a data selection method of an electronic device according to various exemplary embodiments of the present disclosure. Figure 4A 2 is a diagram illustrating an example of a detection object according to various exemplary embodiments of the present disclosure. Figure 4B is a diagram describing a distance map for each object according to various exemplary embodiments of the present disclosure. Figure 4C is a diagram describing weight determination according to various exemplary embodiments of the present disclosure.

[0070] In S100, the processor 140 of the electronic device 100 may receive an image from the camera 110. The camera 110 may transmit to the processor 140 an image obtained by capturing the periphery of the camera 110. The processor 140 may receive the image transmitted from the camera 110.

[0071] In S110, the processor 140 may detect at least one object (eg, a vehicle) from the image. The object may be a vehicle. Figure 4A , the processor 140 may detect the first object 410 and the second object 420 from the image in front of the vehicle obtained by the camera 110. At this time, the processor 140 may detect the objects from the image using a well-known object detection technology.

[0072] In S120, the processor 140 may refer to the lookup table stored in the memory 130 to map the distance (or object distance) to the detected object. Figure 4B , the processor 140 may search for a first frame size 430 that is most similar to the size of the first object 410, and may map the average value of the distance 7.7 m corresponding to the found first frame size 430 to the distance of the first object 410. In addition, the processor 140 may search for a second frame size 440 that is most similar to the size of the second object 420, and may map the average value of the distance 87.5 m corresponding to the found second frame size 440 to the distance of the second object 420.

[0073] In S130, the processor 140 may be configured to determine the weight based on the mapping distance. The processor 140 may verify the target rate and the current rate corresponding to the mapping distance. The target rate may be preset by a system designer or user. The processor 140 may apply the target rate and the current rate to the above equation 1 to determine the weight. Figure 4C , when each of the data acquisition target rate and the data acquisition current rate corresponding to the mapping distance 7.7 m of the first object 410 is "1", the processor 140 may be configured to determine the weight 2 (=1 / 6*12 / 1) using the target rate and the current rate. When the mapping distance is 3.85 m, since the target rate and the current rate corresponding to the mapping distance are 4 and 10, respectively, the processor 140 may be configured to determine the weight 0.8 (=4 / 6*12 / 10).

[0074] In S140, the processor 140 may determine the entropy by using the determined weight. The processor 140 may be configured to determine the entropy of each object using Equation 2 above.

[0075] In S150 , the processor 140 may be configured to determine whether the determined entropy is greater than or equal to a reference value.

[0076] When it is determined that the determined entropy is greater than or equal to the reference value, in S160 , the processor 140 may select the data of the object as data required to generate the GT data.

[0077] Figure 5 2 is a diagram for describing the effects of a data selection method according to various exemplary embodiments of the present disclosure.

[0078] The data selection method according to various exemplary embodiments of the present disclosure may customize the distance weight for a predetermined area of ​​interest (e.g., 20m to 90m). For example, when the detection target of the rear side view camera is to detect an object within a distance of 20m to 90m, the electronic device 100 may apply a greater distance weight to the object within a distance of 20m to 90m to select the object data.

[0079] See also Figure 5 , it can be verified that active learning using distance-weighted entropy Entropy 2 includes more long-range object data than existing active learning random and entropy-based active learning Entropy 1.

[0080] Figure 6 is a block diagram illustrating a computing system for executing a data selection method according to various exemplary embodiments of the present disclosure.

[0081] See also Figure 6 , the computing system 1000 may include at least one processor 1100 , a memory 1300 , a user interface input device 1400 , a user interface output device 1500 , a storage device 1600 , and a network interface 1700 connected to each other via a bus 1200 .

[0082] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes instructions stored in the memory 1300 and / or the storage device 1600. The memory 1300 and the storage device 1600 may include various types of volatile or non-volatile storage media. For example, the memory 1300 may include a read-only memory (ROM) 1310 and a random access memory (RAM) 1320.

[0083] Thus, the operation of the method or algorithm described in conjunction with the exemplary embodiments included in the specification may be directly implemented with a hardware module, a software module, or a combination of a hardware module and a software module, which is executed by the processor 1100. The software module may reside on a storage medium (i.e., memory 1300 and / or storage device 1600) such as RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, removable disk, and CD-ROM. An exemplary storage medium may be coupled to the processor 1100. The processor 1100 may read information from the storage medium and may write information to the storage medium. Alternatively, the storage medium may be integrated with the processor 1100. The processor 110 and the storage medium may reside in an application specific integrated circuit (ASIC). The ASIC may reside in a user terminal. In another case, the processor 1100 and the storage medium may reside in a user terminal as separate components.

[0084] Embodiments of the present disclosure may obtain object data for each distance and object data for each category in a balanced manner.

[0085] Additionally, various exemplary embodiments of the present disclosure may select a data acquisition rate for a desired distance.

[0086] Furthermore, various exemplary embodiments of the present disclosure may variably adjust weights based on a desired data acquisition amount to generate adaptive weights.

[0087] In various exemplary embodiments of the present disclosure, each of the operations described above may be performed by a control device, and the control device may be configured by a plurality of control devices or an integrated single control device.

[0088] In various exemplary embodiments of the present disclosure, the memory and the processor may be provided as one chip, or may be provided as separate chips.

[0089] In different exemplary embodiments of the present disclosure, the scope of the present disclosure includes software or machine-executable commands (e.g., operating systems, applications, firmware, programs, etc.) for enabling operations of methods according to different embodiments to be executed on a device or computer, and non-transitory computer-readable media including such software or commands stored thereon and executable on a device or computer.

[0090] In various exemplary embodiments of the present disclosure, the control device may be implemented in the form of hardware or software, or may be implemented in a combination of hardware and software.

[0091] Software implementations may include software components (or elements), object-oriented software components, class components, task components, procedures, functions, properties, processes, subroutines, program code segments, drivers, firmware, microcodes, data, databases, data structures, tables, arrays, and variables. Software, data, etc. may be stored in a memory and executed by a processor. The memory or processor may be implemented in various ways known to those skilled in the art, including common knowledge in the art.

[0092] Furthermore, terms such as “unit”, “module” and the like included in the specification mean a unit for processing at least one function or operation, which can be implemented by hardware, software or a combination thereof.

[0093] In the flowcharts described with reference to the accompanying drawings, the flowcharts may be executed by a controller or a processor. The order of the operations in the flowcharts may be changed, a plurality of operations may be combined, or any operation may be divided, and a predetermined operation may not be performed. In addition, the operations in the flowcharts may be performed sequentially, but not necessarily sequentially. For example, the order of the operations may be changed, and at least two operations may be performed in parallel.

[0094] Hereinafter, the fact that a plurality of hardwares are operably coupled may include the fact that a direct and / or indirect connection between the plurality of hardwares is established by wire and / or wirelessly.

[0095] In an exemplary embodiment of the present disclosure, a vehicle may be referred to as being based on a concept including various vehicles. In some cases, a vehicle may be interpreted as being based on a concept including not only various land vehicles traveling on roads (such as cars, motorcycles, trucks, and buses) but also various vehicles such as airplanes, drones, ships, etc.

[0096] For ease of explanation and accurate definition of the appended claims, the terms "up", "down", "inside", "outside", "upward", "downward", "front", "back", "back", "interior", "exterior", "inward", "outward", "interior", "exterior", "inside", "outside", "forward", and "rearward" are used to describe features of the exemplary embodiments with reference to the locations of such features shown in the drawings. It should be further understood that the term "connected" or its derivatives refer to both direct and indirect connections.

[0097] The term "and / or" may include any combination of the multiple related listed items or the multiple related listed items. For example, "A and / or B" includes all three cases, such as "A", "B" and "A and B".

[0098] In an exemplary embodiment of the present disclosure, “at least one of A and B” may refer to “at least one of A or B” or “at least one of a combination of at least one of A and B”. Furthermore, “one or more of A and B” may refer to “one or more of A or B” or “one or more of a combination of one or more of A and B”.

[0099] In this specification, unless otherwise stated, a singular expression includes a plural expression unless the context clearly indicates otherwise.

[0100] In the exemplary embodiments of the present disclosure, it should be understood that terms such as “including” or “having” are intended to specify the presence of features, quantities, steps, operations, elements, parts, or a combination thereof described in the specification, and do not exclude the possibility of adding or existing one or more other features, quantities, steps, operations, elements, parts, or a combination thereof.

[0101] According to the exemplary embodiment of the present disclosure, components may be combined with each other to be implemented as one, or some components may be omitted.

[0102] For the purpose of illustration and description, the foregoing descriptions of specific exemplary embodiments of the present disclosure have been presented. They are not intended to be exhaustive or to limit the present disclosure to the precise form disclosed, and it is apparent that many modifications and variations are feasible in light of the above teachings. In order to illustrate certain principles of the present invention and their practical applications, exemplary embodiments have been selected and described to enable other persons skilled in the art to carry out and utilize various exemplary embodiments of the present disclosure and various substitutions and modifications thereof. The scope of the present disclosure is intended to be limited by the appended claims and their equivalents.

Claims

1. An electronic device, comprising: a camera configured to obtain an image; as well as a processor operatively connected to the camera and configured to detect an object from the image, Wherein, the processor is configured to: estimating an object distance of a detected object; Based on the estimated object distance, a weight is determined; By using determined weights, entropy is determined; as well as Based on the determined entropy, it is determined whether to obtain data of the detected object.

2. The electronic device according to claim 1, wherein: The processor is further configured to: Referring to a lookup table, searching for a box size that is most similar to the size of the detected object; as well as The distance mapped to the found box size is estimated as the object distance of the detected object.

3. The electronic device according to claim 2, wherein: The processor is further configured to: matching and classifying, for each predetermined box size, an object having a size most similar to a previously obtained object; For each box size, determine the average distance of the classified objects; as well as Based on the frame size, the determined average distance is determined as the object distance.

4. The electronic device according to claim 3, further comprising: A memory is operatively connected to the processor and stores the lookup table in which the object distance is defined according to the frame size.

5. The electronic device according to claim 1, wherein: The processor is further configured to: setting a data acquisition target rate corresponding to the estimated object distance; verifying a current rate of data acquisition corresponding to the estimated object distance; and The weight is determined using the data acquisition target rate and the data acquisition current rate.

6. The electronic device according to claim 5, wherein: The processor is further configured to determine the entropy by using the determined weights and a probability value of a probability that the detected object will belong to a predetermined category.

7. The electronic device according to claim 1, wherein: The processor is further configured to: determining whether the entropy is greater than or equal to a predetermined reference value; as well as Based on the conclusion that the entropy is greater than or equal to the reference value, data of the detected object corresponding to the entropy is selected as data required to generate ground truth data.

8. The electronic device according to claim 1, wherein: The processor is also configured to detect the object from the image using a deep learning network.

9. The electronic device according to claim 8, wherein: The deep learning network outputs a probability value for the probability that the detected object will belong to a predetermined category.

10. A data selection method for an electronic device, the data selection method comprising: Acquire images using a camera; detecting an object from the image by a processor operatively connected to the camera; estimating, by the processor, an object distance of the detected object; determining, by the processor, a weight based on the estimated object distance; determining, by the processor, entropy by using the determined weights; as well as The processor determines whether to obtain data of the detected object based on the determined entropy.

11. The data selection method according to claim 10, wherein: Estimating the object distance of the detected object comprises: Referring to a lookup table, searching for a box size that is most similar to the size of the detected object; and The distance mapped to the found box size is estimated as the object distance of the detected object.

12. The data selection method according to claim 11, further comprising: matching and classifying, for each predetermined box size, an object having a size most similar to a previously obtained object; For each box size, determine the average distance of the classified objects; as well as Based on the frame size, the determined average distance is determined as the object distance.

13. The data selection method according to claim 12, further comprising: generating the lookup table using the object distance according to the frame size; as well as The look-up table is stored in a memory operatively connected to the processor.

14. The data selection method according to claim 10, wherein: Determining the weight includes: setting a data acquisition target rate corresponding to the estimated object distance; verifying a current rate of data acquisition corresponding to the estimated object distance; and The weight is determined using the data acquisition target rate and the data acquisition current rate.

15. The data selection method according to claim 14, wherein: Determining the entropy includes determining the entropy by using the determined weights and a probability value of a probability that the detected object will belong to a predetermined category.

16. The data selection method according to claim 10, wherein: Determining whether to obtain the data of the object includes: determining whether the entropy is greater than or equal to a predetermined reference value; and Based on the conclusion that the entropy is greater than or equal to the reference value, data of the detected object corresponding to the entropy is selected as data required to generate ground truth data.

17. The data selection method according to claim 10, wherein: Detecting the data includes: The object is detected from the image using a deep learning network.

18. The data selection method according to claim 17, wherein: Detecting the data includes: A probability value of the probability that the detected object will belong to a predetermined category is outputted by the deep learning network.

Images