Method, Apparatus and Computer-readable Medium for Identifying Foreign Substances in Food Using Terahertz Radar

Abstract

A method for identifying foreign substances in food using terahertz radar according to one embodiment of the present disclosure comprises: transmitting a terahertz radar signal toward a food sample and receiving a terahertz radar signal that passes through or is reflected from the food sample to generate a first radar image; generating a second radar image by removing speckle noise from the first radar image; generating a third radar image by applying multiple thresholds to the gradient of signal intensity in the second radar image; and identifying foreign substances contained in the food sample based on the third radar image.

Description

Technology Field

[0001] The present disclosure relates to a method, apparatus, and computer-readable recording medium for identifying foreign substances in food using terahertz radar. Background Technology

[0002] THz radar corresponds to a frequency range of 0.1 to 10 THz in the electromagnetic spectrum, and is located between microwaves and infrared. Additionally, the wavelength range of THz radar can be 30 μm to 3 mm, so THz radar can be the shortest wavelength radio wave and the longest wavelength light wave.

[0003] Therefore, THz radar can simultaneously possess the spatial resolution capability of light waves and the material penetration capability of radio waves. Furthermore, due to its low ionization energy, THz radar has characteristics that make it relatively safe for the human body and may have a unique absorption spectrum.

[0004] As THz spectroscopy and imaging technologies utilizing such material penetration and unique spectra advance, attempts are being made to apply them in various fields such as biology, medicine, and food. In particular, THz radar technology is attracting attention for non-destructively detecting soft substances, such as plastics or insects, which were difficult to identify with existing inspection systems, without altering the state of the food.

[0005] The aforementioned background technology is technical information that the inventor possessed for the derivation of the present invention or acquired during the process of deriving the present invention, and it cannot be considered as prior art disclosed to the general public prior to the filing of the present invention. The problem to be solved

[0006] Some embodiments according to the present disclosure aim to provide a method, apparatus, and computer-readable recording medium for identifying foreign substances in food using terahertz radar. The problems to be solved by the present invention are not limited to those mentioned above, and other problems and advantages of the present invention not mentioned can be understood from the following description and will be more clearly understood by embodiments of the present invention. Furthermore, it will be seen that the problems and advantages to be solved by the present invention can be realized by means and combinations thereof as set forth in the claims. means of solving the problem

[0007] As a technical means for achieving the technical problem described above, the first aspect of the present disclosure may provide a method for identifying foreign substances in food using terahertz radar, comprising: transmitting a terahertz radar signal toward a food sample and receiving a terahertz radar signal that passes through or is reflected from the food sample to generate a first radar image; generating a second radar image by removing speckle noise from the first radar image; generating a third radar image by applying multiple thresholds to the gradient of signal intensity in the second radar image; and identifying foreign substances contained in the food sample based on the third radar image.

[0008] A second aspect of the present disclosure provides a device for identifying foreign substances in food using terahertz radar, comprising: at least one memory; and at least one processor; wherein the processor transmits a terahertz radar signal toward a food sample and receives a terahertz radar signal that passes through or is reflected from the food sample to generate a first radar image, generates a second radar image by removing speckle noise from the first radar image, generates a third radar image by applying multiple thresholds to the slope of the signal intensity in the second radar image, and identifies foreign substances contained in the food sample based on the third radar image.

[0009] A third aspect of the present disclosure may provide a computer-readable recording medium having a program for executing a method according to a first aspect on a computer.

[0010] In addition to this, other methods for implementing the present invention, other systems, and computer-readable recording media storing a computer program for executing said methods may be further provided.

[0011] Other aspects, features, and advantages other than those described above will become clear from the following drawings, claims, and detailed description of the invention. Brief explanation of the drawing

[0012] FIG. 1 is a diagram illustrating an example of a system for identifying foreign substances in food using terahertz radar according to one embodiment. FIG. 2 is a drawing for illustrating an example of the internal configuration of a terahertz radar device according to one embodiment. FIG. 3 is a flowchart illustrating an example of a method for identifying foreign substances in food using terahertz radar according to one embodiment. FIG. 4 is a flowchart illustrating another example of a method for identifying foreign substances in food using terahertz radar according to one embodiment. FIG. 5 is a drawing illustrating another example of the internal configuration of a terahertz radar device according to one embodiment. FIGS. 6a and 6b are drawings for illustrating examples of radar images according to one embodiment. Specific details for implementing the invention

[0013] The advantages and features of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments presented below, but can be implemented in various different forms and should be understood to include all modifications, equivalents, and substitutions that fall within the spirit and scope of the present invention. The embodiments presented below are provided to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention. In describing the present invention, detailed descriptions of related known technologies are omitted if it is determined that such detailed descriptions may obscure the essence of the present invention.

[0014] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as “comprising” or “having” are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0015] Some embodiments of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented by various numbers of hardware and / or software configurations that execute specific functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors or by circuit configurations for a specific function. Additionally, for example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented as algorithms executed on one or more processors. Furthermore, the present disclosure may employ prior art for electronic configuration, signal processing, and / or data processing, etc. Terms such as “mechanism,” “element,” “means,” and “configuration” may be used broadly and are not limited to mechanical and physical configurations.

[0016] Furthermore, the connecting lines or connecting members between the components depicted in the drawings are merely illustrative of functional connections and / or physical or circuit connections. In the actual device, connections between components may be represented by various alternative or added functional connections, physical connections, or circuit connections.

[0017] Embodiments are described in detail below with reference to the attached drawings. However, embodiments may be implemented in various different forms and are not limited to the examples described herein.

[0018] FIG. 1 is a drawing for explaining an example of a system for identifying foreign substances in food using terahertz radar according to one embodiment, and FIG. 2 is a drawing for explaining an example of the internal configuration of a terahertz radar device according to one embodiment.

[0019] Referring to FIG. 1, a foreign substance identification system (1) (hereinafter referred to as the 'system') using terahertz radar in food may be configured to include a terahertz radar device (10) and a sample transfer device (30).

[0020] A terahertz radar device (10) can be implemented to include various components for performing the task of identifying foreign substances using terahertz radar with respect to a sample (21)(22), which is an object to be examined for whether foreign substances are included. For example, the terahertz radar device (10) can radiate terahertz radar over the entire sample by scanning the sample (21)(22), and identify foreign substances using radar signals received from the sample (21)(22).

[0021] The sample transport device (30) is a component that transports a test object. In the present disclosure, the test object may be a food sample (21)(22). For example, the sample transport device (30) may be implemented in a form that includes a conveyor belt, and the food sample (21)(22) may be placed on the upper part of the conveyor belt and transported in the horizontal axis direction of the sample transport device (30). At this time, the terahertz radar device (10) may transmit terahertz radar (or terahertz waves) to the upper part of the test object being transported on the upper part of the conveyor belt. For example, the food sample may be transported from the location of the food sample (21) to the location of the food sample (22) and undergo an inspection for the presence of foreign substances.

[0022] According to one embodiment of the present disclosure, a terahertz radar device (200) may include a radar transceiver (210), a radar image generation unit (220), and a foreign substance identification unit (230).

[0023] The radar transceiver (210) is a component that transmits a terahertz radar signal toward a food sample and receives a terahertz radar signal that passes through or is reflected from the food sample. More specifically, the radar transceiver (210) may include a radar transmitter (not shown) and a radar receiver (not shown).

[0024] The radar transmitter is a component that generates and supplies terahertz radar, and can be replaced with various terms such as terahertz radar (terahertz wave) light source, terahertz radar (terahertz wave) source, etc. Here, terahertz wave refers to electromagnetic waves in the terahertz region, and preferably may have a frequency of 0.1 THz to 10 THz. However, it goes without saying that even if it falls outside this range, a region that does not deviate significantly from it may be recognized as a terahertz wave in the present disclosure.

[0025] A radar receiver is a component that collects and detects terahertz radar incident on a food sample. For example, the radar receiver can receive terahertz radar signals transmitted through or reflected from the food sample. Meanwhile, in the present disclosure, the radar receiver may be implemented in a form in which a reflection detection module that detects a reflected signal and a transmission detection module that detects a transmitted signal are implemented independently.

[0026] The radar image generation unit (220) can generate a radar image using a terahertz radar signal received by the radar transceiver (210).

[0027] For example, the radar image generation unit (220) may be configured to include an amplifier (not shown), an AD converter (not shown), and a signal processing module (not shown). A radar signal received by the radar transceiver (210) may be amplified through the amplifier, then converted into a digital signal by the AD converter and transmitted to the signal processing module. The signal processing module may generate a two-dimensional radar image by converting the transmitted signal into pixel values ​​corresponding to image (or image) coordinates through software.

[0028] Meanwhile, the user can visually observe the generated 2D radar image to determine in real time whether foreign substances are contained in the food.

[0029] Alternatively, the foreign substance identification unit (230) can identify foreign substances contained in the food sample based on a two-dimensional radar image.

[0030] In other words, the quality of the radar image can be an important factor in identifying foreign substances within a food sample. To this end, the present disclosure describes in detail a method for generating a radar image in which foreign substances can be clearly identified.

[0031] The radar image generation unit (220) can generate a first radar image based on a terahertz radar signal transmitted through or reflected from a food sample. Here, the first radar image refers to an original image.

[0032] According to one embodiment, the radar image generation unit (220) can generate a two-dimensional radar image based on the I-channel signal strength and Q-channel signal strength of the terahertz radar signal received by the radar transceiver (210). The I-channel signal of the received signal represents the real part of the complex signal. Also, the Q-channel signal of the received signal represents the imaginary part of the complex signal. For example, in a two-dimensional radar image (e.g., a complex plane), the I-channel signal strength of the received signal can be represented on the horizontal axis, and the Q-channel signal strength of the received signal can be represented on the vertical axis.

[0033] The radar image generation unit (220) can generate a second radar image by removing speckle noise from the first radar image. Speckle noise refers to noise caused by light interference phenomena caused by fine irregularities on the surface of an object.

[0034] According to one embodiment, the radar image generation unit (220) can remove speckle noise in the first radar image by comparing the magnitude of the signal component of speckle noise included in the first radar image with the magnitude of an adjacent other signal component. That is, the radar image generation unit (220) can remove speckle noise by performing smoothing on the first radar image. For example, techniques for performing smoothing may include a moving average technique, a Gaussian smoothing technique, a median filter technique, a low pass filter technique, etc.

[0035] Here, the radar image generation unit (220) can determine the size of the smoothing window based on the extent to which the signal component of speckle noise is included in the first radar image. Here, the smoothing window refers to an area that includes other adjacent signal components that compare the magnitude of the signal component with respect to the speckle noise. For example, depending on the extent to which the signal component of speckle noise is included in the first radar image, the size of the smoothing window may be determined to be 8 pixels, 16 pixels, etc.

[0036] The radar image generation unit (220) can generate a third radar image by applying multiple thresholds to the gradient of the signal strength in the second radar image.

[0037] More specifically, the radar image generation unit (220) can calculate the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the second radar image. Additionally, the radar image generation unit (220) can remove the local non-maximum value of the slope of the signal strength in a predetermined area included in the second radar image based on the calculated slope of the signal strength. The radar image generation unit (220) can generate a two-dimensional radar image in which the slope of the signal strength is expressed differently depending on which of the multiple thresholds the magnitude of the slope of the signal strength is greater than or equal to.

[0038] In other words, in the third radar image to which multiple thresholds are applied to the slope of the signal intensity, the interface between the food and foreign substances within the food can be represented more clearly. Here, the sensitivity to the slope in the third radar image can be determined (adjusted) depending on the interval between the thresholds included in the multiple thresholds.

[0039] According to one embodiment, the radar image generation unit (220) can obtain a radar image in the frequency domain through frequency transformation (e.g., Fast Fourier Transform) for the second radar image. The radar image generation unit (220) can remove noise components from the radar image in the frequency domain using a noise removal filter, and can obtain a restored radar image (spatial domain) through frequency inverse transformation of the radar image in the frequency domain from which noise components have been removed. Additionally, the radar image generation unit (220) can calculate the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the restored radar image.

[0040] According to one embodiment, the multiple threshold may include a plurality of thresholds determined based on the rate of change of signal strength regarding the second radar image in the spatial domain and the rate of change of phase regarding the radar image in the frequency domain. In other words, the multiple threshold may be determined by further considering the rate of change of phase regarding the radar image in the frequency domain in addition to the rate of change of signal strength regarding the second radar image in the spatial domain.

[0041] According to one embodiment, the foreign substance identification unit (230) may determine a point where the slope of the signal strength in the third radar image is greater than or equal to a preset reference value as the boundary point between the area containing foreign substances and the area not containing foreign substances within the food. Here, the preset reference value may be determined as any one of the threshold values ​​included in multiple thresholds.

[0042] According to another embodiment, the foreign substance identification unit (230) inputs the third radar image into a deep learning model that has been trained using the third radar image, and can obtain an identification result for foreign substances contained in a food sample as output data of the deep learning model. Here, the deep learning model may be composed of an input layer, a hidden layer, and an output layer, and each layer may be composed of multiple neurons. In addition, each neuron can calculate an output value by applying a value obtained by multiplying an input value by a weight to an activation function, and each layer transmits a signal to the next layer, and the signal can be controlled through weights and biases. That is, the deep learning model optimizes weights and biases using training data, and thereby can learn the complex relationship between input and output. Meanwhile, the deep learning model can be implemented as various types of models such as a Multi-Layer Perceptron (MLP), a Convolutional Neural Network (CNN), and a Recurrent Neural Network (RNN). For example, a deep learning model can be trained to classify whether foreign substances are included or to classify foreign substances included in food by using a third radar image as training data. The deep learning model, once trained, can use the third radar image as input data and output the identification result of foreign substances included in the food sample as output data.

[0043] FIG. 3 is a flowchart illustrating an example of a method for identifying foreign substances in food using terahertz radar according to one embodiment.

[0044] Referring to FIG. 3, the method for identifying foreign substances in food using terahertz radar may include steps 310 to 370. However, in addition to this, operations described as being performed by a terahertz radar device with reference to FIG. 1 and FIG. 2 may also be included in the method for identifying foreign substances in food using terahertz radar.

[0045] First, in step 310, the terahertz radar device can transmit a terahertz radar signal toward a food sample and receive a terahertz radar signal that passes through or is reflected from the food sample to generate a first radar image.

[0046] For example, a terahertz radar device can generate a two-dimensional radar image based on the I-channel signal strength and Q-channel signal strength of a received terahertz radar signal.

[0047] In step 330, the terahertz radar device can generate a second radar image by removing speckle noise from the first radar image.

[0048] For example, a terahertz radar device can generate a two-dimensional radar image with speckle noise removed by comparing the magnitude of a signal component of speckle noise included in a first radar image with the magnitude of an adjacent other signal component.

[0049] In addition, the terahertz radar device can determine the size of the smoothing window based on the extent to which the signal component of speckle noise is included in the first radar image.

[0050] In step 350, the terahertz radar device can generate a third radar image by applying multiple thresholds to the gradient of signal strength in the second radar image.

[0051] For example, a terahertz radar device can calculate the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the second radar image, remove local non-maximum values ​​of the slope of the signal strength in a predetermined area included in the second radar image, and apply multiple thresholds to the slope of the signal strength to generate a two-dimensional radar image in which the slope of the signal strength is expressed differently depending on which of the multiple thresholds the magnitude of the slope of the signal strength is greater than or equal to.

[0052] In addition, the terahertz radar device can obtain a radar image in the frequency domain through frequency conversion of the second radar image, and can remove noise components from the radar image in the frequency domain using a noise removal filter. In addition, the terahertz radar device can obtain a restored radar image through frequency inverse conversion of the radar image in the frequency domain from which noise components have been removed, and can calculate the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the restored radar image.

[0053] Meanwhile, according to one embodiment, the multiple thresholds may include a plurality of thresholds determined based on the rate of change of signal strength regarding the second radar image in the spatial domain and the rate of change of phase regarding the radar image in the frequency domain.

[0054] Referring again to FIG. 3, in step 370, the terahertz radar device can identify foreign substances contained in the food sample based on the third radar image.

[0055] For example, a terahertz radar device can identify foreign substances by determining a point in a third radar image where the slope of the signal intensity is greater than or equal to a preset reference value as the boundary point between the area containing the foreign substance and the area not containing the foreign substance within the food.

[0056] As another example, a terahertz radar device can identify foreign substances by inputting a third radar image into a deep learning model that has been trained using the third radar image, and obtaining an identification result of foreign substances contained in a food sample as the output data of the deep learning model.

[0057] FIG. 4 is a flowchart illustrating another example of a method for identifying foreign substances in food using terahertz radar according to one embodiment.

[0058] First, in step 410, the terahertz radar device can generate a second radar image from which speckle noise in the first radar image has been removed. In this regard, the description with reference to FIG. 3 can be applied in the same way to step 410.

[0059] Likewise, in step 420, the terahertz radar device can acquire a radar image in the frequency domain through frequency conversion of the second radar image, and in step 430, the terahertz radar device can remove noise components from the radar image in the frequency domain using a noise removal filter.

[0060] Subsequently, in step 440, the terahertz radar device can calculate the rate of change of phase regarding the radar image in the frequency domain, and in step 450, the terahertz radar device can calculate the rate of change of signal strength regarding the second radar image in the space domain.

[0061] In step 460, the terahertz radar device can determine a plurality of multiple thresholds based on the rate of change of signal strength for the second radar image in the spatial domain and the rate of change of phase for the radar image in the frequency domain.

[0062] Subsequently, in step 470, a third radar image can be generated by applying multiple thresholds to the slope of the signal strength in the second radar image.

[0063] FIG. 5 is a drawing illustrating another example of the internal configuration of a terahertz radar device according to one embodiment.

[0064] Referring to FIG. 5, the terahertz radar device (500) includes a processor (510), memory (520), an input / output interface (530), and a communication module (540). For convenience of explanation, FIG. 5 only illustrates components related to the present invention. Accordingly, other general-purpose components in addition to those illustrated in FIG. 5 may be included in the terahertz radar device (500). Furthermore, it is obvious to those skilled in the art that the processor (510), memory (520), input / output interface (530), and communication module (540) illustrated in FIG. 5 may be implemented as independent devices.

[0065] Meanwhile, at least some of the operations described as being performed by a terahertz radar device with reference to FIGS. 1 to 4 may be performed by a terahertz radar device (500).

[0066] The processor (510) can process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. Here, instructions may be provided from memory (520) or an external device. Additionally, the processor (510) can control the overall operation of other components included in the terahertz radar device (500).

[0067] The processor (510) can generate a first radar image by transmitting a terahertz radar signal toward a food sample and receiving a terahertz radar signal that passes through or is reflected from the food sample.

[0068] For example, the processor (510) can generate a two-dimensional radar image based on the I-channel signal strength and Q-channel signal strength of the received terahertz radar signal.

[0069] The processor (510) can generate a second radar image by removing speckle noise from the first radar image.

[0070] For example, the processor (510) can generate a two-dimensional radar image with speckle noise removed by comparing the magnitude of the signal component of speckle noise included in the first radar image with the magnitude of another adjacent signal component.

[0071] Additionally, the processor (510) can determine the size of the smoothing window based on the extent to which the signal component of speckle noise is included in the first radar image.

[0072] The processor (510) can generate a third radar image by applying multiple thresholds to the gradient of the signal strength in the second radar image.

[0073] For example, the processor (510) can calculate the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the second radar image, remove the local non-maximum value of the slope of the signal strength in a predetermined area included in the second radar image, and apply multiple thresholds to the slope of the signal strength to generate a two-dimensional radar image in which the slope of the signal strength is expressed differently depending on which of the multiple thresholds the magnitude of the slope of the signal strength is greater than or equal to.

[0074] In addition, the processor (510) can obtain a radar image in the frequency domain through frequency conversion of the second radar image and can remove noise components from the radar image in the frequency domain using a noise removal filter. Additionally, the terahertz radar device can obtain a restored radar image through frequency inverse conversion of the radar image in the frequency domain from which noise components have been removed, and can calculate the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the restored radar image.

[0075] Meanwhile, according to one embodiment, the multiple thresholds may include a plurality of thresholds determined based on the rate of change of signal strength regarding the second radar image in the spatial domain and the rate of change of phase regarding the radar image in the frequency domain.

[0076] The processor (510) can identify foreign substances contained in the food sample based on the third radar image.

[0077] For example, the processor (510) can identify the foreign substance by determining a point in the third radar image where the slope of the signal strength is greater than or equal to a preset reference value as the boundary point between the area containing the foreign substance and the area not containing the foreign substance within the food.

[0078] As another example, the processor (510) can identify foreign substances by inputting the third radar image into a deep learning model that has been trained using the third radar image, and obtaining an identification result of foreign substances contained in a food sample as output data of the deep learning model.

[0079] The processor (510) may be implemented as an array of multiple logic gates, or as a combination of a general-purpose microprocessor and memory storing a program that can be executed on the microprocessor. For example, the processor (510) may include a general-purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, etc. In some environments, the processor (510) may include an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), etc. For example, the processor (510) may refer to a combination of processing devices such as a combination of a digital signal processor (DSP) and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors combined with a digital signal processor (DSP) core, or any other combination of such configurations.

[0080] The memory (520) may include any non-transient computer-readable recording medium. As an example, the memory (520) may include a non-perishable permanent mass storage device such as RAM (random access memory), ROM (read only memory), disk drive, SSD (solid state drive), flash memory, etc. As another example, a non-perishable permanent mass storage device such as ROM, SSD, flash memory, disk drive, etc. may be a separate permanent storage device distinct from the memory. Additionally, the memory (520) may store an operating system (OS) and at least one program code (e.g., code for the processor (510) to perform the operation described above with reference to FIGS. 1 to 4).

[0081] These software components may be loaded from a computer-readable recording medium separate from the memory (520). This separate computer-readable recording medium may be a recording medium that can be directly connected to the terahertz radar device (500), and may include, for example, a computer-readable recording medium such as a floppy drive, disk, tape, DVD / CD-ROM drive, memory card, etc. Alternatively, the software components may be loaded into the memory (520) via a communication module (540) that is not a computer-readable recording medium. For example, at least one program may be loaded into the memory (520) based on a computer program (e.g., a computer program for the processor (510) to perform the operation described above with reference to FIGS. 1 to 4) which is installed by files provided through the communication module (540) by developers or a file distribution system that distributes installation files of the application.

[0082] The input / output interface (530) may be a means for interfacing with a device for input or output (e.g., keyboard, mouse, etc.) that may be connected to or included in the terahertz radar device (500). In FIG. 5, the input / output interface (530) is shown as an element configured separately from the processor (510), but is not limited thereto, and the input / output interface (530) may be configured to be included in the processor (510).

[0083] The communication module (540) may provide a configuration or function for the terahertz radar device (500) to communicate with an external device through a network. Additionally, the communication module (540) may provide a configuration or function for the terahertz radar device (500) to communicate with another external device. For example, control signals, commands, data, etc. provided under the control of the processor (510) may be transmitted to an external device through the communication module (540) and the network.

[0084] Meanwhile, although not shown in FIG. 5, the terahertz radar device (500) may further include a display device. Alternatively, the terahertz radar device (500) may be connected to an independent display device via wired or wireless communication to transmit and receive data to and from each other.

[0085] FIGS. 6a and 6b are drawings for illustrating examples of radar images according to one embodiment.

[0086] Referring to FIGS. 6a and 6b, a terahertz radar device can generate a first radar image (611) (612) by receiving a terahertz radar signal transmitted through or reflected from a food sample. The first radar image (612) of FIG. 6b is an image with a relatively high speckle noise component compared to the first radar image (611) of FIG. 6a.

[0087] A terahertz radar device can generate a second radar image (621)(622) by removing speckle noise from a first radar image (611)(612). For example, the terahertz radar device can remove speckle noise from the first radar image (611)(612) by performing a flattening process.

[0088] A terahertz radar device can generate a third radar image (631)(632) by applying multiple thresholds to the slope of the signal strength in the second radar image (621)(622). Looking at the third radar image (631) of FIG. 6a and the third radar image (632) of FIG. 6b, it can be seen that the boundary between the food sample and the foreign substance can be clearly distinguished.

[0089] Unless explicitly stated or contrary to the order of the steps constituting the method according to the present invention, said steps may be performed in a suitable order. The present invention is not necessarily limited by the order in which said steps are described. The use of all examples or exemplary terms (e.g., etc.) in the present invention is merely for the purpose of describing the present invention in detail, and the scope of the present invention is not limited by said examples or exemplary terms unless limited by the claims. Furthermore, those skilled in the art will understand that various modifications, combinations, and changes may be made according to design conditions and factors within the scope of the claims or equivalents to which they are added.

[0090] Accordingly, the scope of the present invention should not be limited to the embodiments described above, and all scopes equivalent to or equivalently modified from the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the concept of the present invention.

Claims

Claim 1 A method for identifying foreign substances in food using terahertz radar, comprising: transmitting a terahertz radar signal toward a food sample and receiving a terahertz radar signal that passes through or is reflected from the food sample to generate a first radar image; generating a second radar image by removing speckle noise from the first radar image; generating a third radar image by applying multiple thresholds to the gradient of signal intensity in the second radar image; and identifying foreign substances contained in the food sample based on the third radar image; wherein the step of identifying foreign substances includes determining a point in the third radar image where the gradient of signal intensity is greater than or equal to a preset reference value as the boundary point between an area containing the foreign substance and an area not containing the foreign substance within the food. Claim 2 A method according to claim 1, wherein the step of generating the first radar image comprises the step of generating a two-dimensional radar image based on the I-channel signal strength and Q-channel signal strength of the received terahertz radar signal. Claim 3 A method according to claim 1, wherein the step of generating the second radar image comprises the step of generating a two-dimensional radar image from which the speckle noise has been removed by comparing the magnitude of the signal component of the speckle noise included in the first radar image with the magnitude of an adjacent other signal component. Claim 4 A method according to claim 3, further comprising the step of determining the size of a smoothing window based on the degree to which the signal component of the speckle noise is included in the first radar image. Claim 5 The method according to claim 1, wherein the step of generating the third radar image comprises: calculating the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the second radar image; removing a local non-maximum value of the slope of the signal strength in a predetermined area included in the second radar image; and generating a two-dimensional radar image in which the slope of the signal strength is expressed differently depending on whether the magnitude of the slope of the signal strength is greater than or equal to any of the multiple thresholds by applying multiple thresholds to the slope of the signal strength. Claim 6 In claim 5, the step of calculating the slope of the signal strength comprises: a step of obtaining a radar image in the frequency domain through frequency transformation of the second radar image; a step of removing noise components from the radar image in the frequency domain using a noise removal filter; a step of obtaining a restored radar image through frequency inverse transformation of the radar image in the frequency domain from which the noise components have been removed; and a step of calculating the slope of the signal strength for each of the vertical signal component and the horizontal signal component in the restored radar image. Claim 7 A method according to claim 6, wherein the multiple thresholds include a plurality of thresholds determined based on the rate of change of signal strength regarding the second radar image in the spatial domain and the rate of change of phase regarding the radar image in the frequency domain. Claim 8 delete Claim 9 A method according to claim 1, wherein the step of identifying the foreign substance comprises inputting the third radar image into a deep learning model that has been trained using the third radar image, and obtaining an identification result for the foreign substance contained in the food sample as output data of the deep learning model. Claim 10 A device for identifying foreign substances in food using terahertz radar, comprising: at least one memory; and at least one processor; wherein the processor transmits a terahertz radar signal toward a food sample and receives a terahertz radar signal transmitted through or reflected from the food sample to generate a first radar image, generates a second radar image by removing speckle noise from the first radar image, generates a third radar image by applying multiple thresholds to the slope of the signal intensity in the second radar image, and identifies foreign substances contained in the food sample based on the third radar image; wherein the processor determines a point in the third radar image where the slope of the signal intensity is greater than or equal to a preset reference value as the boundary point between an area containing the foreign substance and an area not containing the foreign substance within the food. Claim 11 A computer-readable recording medium having a program for executing the method of claim 1 on a computer.

Images