KR102654020B1 - Stringed instrument value evaluating method and computer program - Google Patents

Abstract

A string instrument valuation method using a string instrument valuation device according to an embodiment of the present invention includes receiving a photographed image of a string instrument and production information of the string instrument; analyzing state information of the string instrument from the captured image; Deriving a basic value of the string instrument based on the production information; deriving a discount value by applying a discount rate to the basic value based on the status information; and visualizing and displaying at least one of the basic value and the discount value of the string instrument. Includes.

Description

Stringed instrument valuation method and computer program {STRINGED INSTRUMENT VALUE EVALUATING METHOD AND COMPUTER PROGRAM}

The present invention relates to a method for evaluating the value of stringed instruments such as violins, violas, cellos, and basses, and to computer programs thereof.

Stringed instruments, such as the violin, viola, cello, and bass, have a long history and tradition, and are of considerable value in themselves, and their value has increased further as famous composers throughout history have created works using stringed instruments.

Each string instrument is handcrafted and made using the best materials and techniques, and in this production method, the skill and effort of the maker are directly related to the value of the string instrument. Additionally, its delicate and complex design and completeness have optimal characteristics for producing excellent sound.

Because stringed instruments have high artistic value, their value tends to increase over time, just like works of art. In the case of the violin, a representative string instrument, the violins used by famous violinists throughout history were later given high value to other players, and some violins are traded for millions of dollars.

However, assessing the value of stringed instruments is not easy. This is because many factors are involved, including the condition of the stringed instrument, the player, the maker, the way it is played, and its history. Therefore, evaluating the value of stringed instruments requires expert evaluation and experience, and requires considerable knowledge, time, and effort.

Meanwhile, financial services with enhanced accessibility are being provided in response to diverse demands for investment methods and technological advancements, and demand for alternative investments that invest in objects other than traditional investment products such as stocks or bonds is increasing. And in order to provide easier access to alternative investments, alternative investments made in an online environment are being developed.

However, currently, the targets of online alternative investments are limited to products for which relatively objective evaluation standards have been established, such as paintings and alcoholic beverages. In the case of expensive used musical instruments such as old string instruments among alternative investment targets, although it is an area with high investment value, factors that make it difficult to accurately evaluate the value of the product compared to other alternative investment products act as a disadvantage, leading to offline auctions rather than online linked investments. There is a problem that investment mainly occurs through the process of .

Korean Patent Publication No. 2022-0167884

The present invention is intended to solve the problems of the prior art described above and to provide a value evaluation method that can accurately and quickly appraise the value of a string instrument.

In addition, the present invention seeks to provide a value evaluation method that can accurately appraise the value of a string instrument by considering damage to each structure of the string instrument and changes in damage over time.

In addition, the present invention seeks to provide a valuation method that can intuitively determine the evaluated value and valuation factors of a string instrument.

However, the technical challenges that this embodiment aims to achieve are not limited to the technical challenges described above, and other technical challenges may exist.

As a means for achieving the above-mentioned technical problem, one embodiment of the present invention provides a method for evaluating the value of a stringed instrument using a stringed instrument value evaluation device, comprising: receiving a photographed image of a stringed instrument and production information of the stringed instrument; analyzing state information of the string instrument from the captured image; Deriving a basic value of the string instrument based on the production information; deriving a discount value by applying a discount rate to the basic value based on the status information; and visualizing and displaying at least one of the basic value and the discount value of the string instrument. may include.

In addition, another embodiment of the present invention provides a method for evaluating the value of a stringed instrument using a stringed instrument valuation device, comprising: receiving a currently captured image of a stringed instrument and production information of the stringed instrument; analyzing current state information of the string instrument from the currently captured image; Deriving a basic value of the string instrument based on the production information; Loading past state information of the string instrument analyzed from a past photographed image generated in the past by photographing the string instrument in the past, and comparing the past state information with the current state information to derive state change information; Deriving price change information by applying the state change information to a regression model; deriving the current value of the string instrument by reflecting the price change information in the basic value; and visualizing and displaying at least one of the basic value and the current value of the string instrument. may include.

The above-described means for solving the problem are merely illustrative and should not be construed as limiting the present invention. In addition to the exemplary embodiments described above, there may be additional embodiments described in the drawings and detailed description of the invention.

According to any one of the means for solving the problems of the present invention described above, it is possible to provide a value evaluation method that can accurately and quickly appraise the value of a string instrument.

In addition, the present invention can provide a value evaluation method that can accurately appraise the value of a string instrument by considering damage to each structure of the string instrument and changes in damage over time.

In addition, the present invention can provide a valuation method that can intuitively know the evaluated value and valuation factors of a string instrument.

1 is a configuration diagram of a string instrument value evaluation system according to an embodiment of the present invention.
Figure 2 is a flowchart illustrating a method for evaluating the value of a string instrument according to an embodiment of the present invention.
Figures 3 and 4 are diagrams for illustrating and illustrating the configuration of a violin in relation to damage information among state information in the method for evaluating the value of a stringed instrument according to an embodiment of the present invention.
Figure 5 is a flowchart illustrating a method for evaluating the value of a string instrument according to an embodiment of the present invention.
Figures 6 and 7 are diagrams illustrating an example of a user interface screen that visualizes and displays a method for evaluating the value of a stringed instrument according to an embodiment of the present invention.

Below, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are given similar reference numerals throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected," but also the case where it is "electrically connected" with another element in between. . In addition, when a part is said to "include" a certain component, this does not mean excluding other components unless specifically stated to the contrary, but may further include other components, and one or more other features. It should be understood that it does not exclude in advance the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

In this specification, 'part' includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more pieces of hardware, and two or more units may be realized using one piece of hardware.

In this specification, some of the operations or functions described as being performed by a terminal or device may instead be performed on a server connected to the terminal or device. Likewise, some of the operations or functions described as being performed by the server may also be performed on a terminal or device connected to the server.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the attached drawings.

1 is a diagram for explaining a string instrument value evaluation system according to an embodiment of the present invention.

Referring to FIG. 1, the stringed instrument valuation system 1 may include a stringed instrument valuation device 100, a network 200, and a user terminal 300.

The stringed instrument valuation device 100 of the stringed instrument valuation system 1 of FIG. 1 is connected to the user terminal 300 and the network 200. For example, as shown in FIG. 1, the financial service providing device 100 may be connected to the user terminal 300 simultaneously or at time intervals. The network 200 refers to a connection structure that allows information exchange between nodes such as terminals and servers, including a local area network (LAN), a wide area network (WAN), and the Internet ( It includes WWW (World Wide Web), wired and wireless data communication networks, telephone networks, and wired and wireless television communication networks. Examples of wireless data communication networks include 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), Wi-Fi, Bluetooth communication, infrared communication, and ultrasonic communication. , Visible Light Communication (VLC), LiFi, etc., but are not limited thereto.

The user terminal 300 is a person who uses the value evaluation service of a string instrument, and provides a captured image of a string instrument to the string instrument value evaluation device 100 through the network 200 or provides the value evaluation result of the string instrument value evaluation device 100. It can be searched through the network 200.

Here, string instruments have their own unique sound and aesthetics and are preferred by many musicians. For example, string instruments produced by Stradivarius and Guarneri, known as famous violins, can be representative examples. These stringed instruments have been well maintained and tuned over a long period of time, and are famous for their unique sounds and beautiful designs. In the case of used stringed instruments, this may include the instrument body and the bow used to string the body. In addition, these string instruments are mostly preferred by professional musicians and are often used in concerts and orchestras.

In the case of expensive string instruments, unlike other expensive investments that have value only as collateral, additional profits can be secured through rental. This is due to the fact that, due to the nature of string instruments, the value of used instruments can be improved through continued use and repair rather than just storage. Therefore, the present invention makes it possible to quickly and accurately determine the value of a stringed instrument in order to invest in an expensive stringed instrument or use a financial service using the stringed instrument as collateral.

Figure 2 is a flowchart illustrating a method of providing financial services using used musical instruments according to an embodiment of the present invention.

Referring to Figure 2, the method for evaluating the value of a string instrument includes a shooting image and production information receiving step (S200), a state information analysis step (S210), a basic value deriving step (S220), a discount value deriving step (S230), and a visualization display step. (S240) may be included.

The stringed instrument valuation method of FIG. 2 may be performed by the stringed instrument valuation device 100 in FIG. 1 . The string instrument valuation device 100 may correspond to a server or computing device that can be connected online with a plurality of user terminals 300 through the network 200, and various data generated in the process of performing the string instrument valuation method. Alternatively, it may include a database (not shown) for storing information. The method for evaluating the value of a stringed instrument according to an embodiment of the present invention may operate in the form of an online platform or application that runs through the device 100 for evaluating the value of a stringed instrument.

The captured image and production information receiving step (S200) may be a step in which the string instrument value evaluation device receives a captured image of the string instrument and production information of the string instrument.

Here, the captured image includes an external image photographed by each part of the exterior of the string instrument and an internal image photographed by each region of the interior of the string instrument, and the internal image is a photograph taken by at least one of endoscopy, X-ray examination, and ultrasound examination. You can.

To evaluate the value of a string instrument, images of the interior as well as the exterior of the string instrument are needed. However, it is generally difficult to photograph the interior of a string instrument by part without disassembling it, and damage may occur during the process of disassembling an old string instrument or opening the front panel, so careful work by an expert is required. Therefore, separate equipment may be required to photograph the interior of a stringed instrument without disassembling it. When using equipment such as the above-mentioned X-ray, endoscope, and ultrasound, internal imaging is possible without disassembling the string instrument.

An endoscope is a flexible camera that allows you to look inside a string instrument (for example, a violin's f-hole). Using an endoscope, it is possible to check the state of damage inside a stringed instrument, and for this purpose, the interior must be sufficiently illuminated through appropriate lighting.

Ultrasonic testing is a method of sending ultrasonic waves through a hole inside a string instrument and measuring the reflected waves to check the internal condition.

X-ray inspection is a way to check the structure and condition of the internal string instrument in more detail. However, X-ray examinations must be performed by experts because of the risk of radiation exposure, and careful consideration must be given before being used to prevent unnecessary damage to stringed instruments.

However, when photographing the interior of a stringed instrument using the above-described photographing device, the image may not be clear and noise may occur compared to an exterior image photographed in a bright space.

Preprocessing of the internal image may be necessary to remove noise occurring in the internal image and obtain a clearer image.

The internal image may be image-processed through at least one of a Gaussian filter, a median filter, a Laplacian filter, and a wavelet filter to remove noise from the captured image.

Here, Gaussian filter may refer to a method of removing noise by smoothing pixel values of an image using Gaussian distribution. Applying a Gaussian filter makes the image softer and reduces noise. In addition, the median filter is a method of finding the median in an image and using it to remove noise. It is effective in removing pixels that are judged to be noise in the image, and is especially useful when noise is distributed throughout the image. can do. And the wavelet filter is a method of decomposing an image using wavelet transform and removing noise from the decomposed image. It removes the high frequency components of the image (irregular patterns such as noise) and removes the low frequency components (static background, etc.) of the image. It can be effective in preserving.

Production information can be entered directly from the user terminal who wishes to evaluate the value of the string instrument. Additionally, production information may be input from information extracted by scanning a string instrument warranty, etc. In another embodiment, the user may separately input information related to the warranty into the string instrument value evaluation device through the user terminal. The warranty-related information entered here may include information that can prove that the warranty was issued by a renowned string instrument expert such as Eric Blot, Carlson & Neumann, and Charles Beare.

The state information analysis step (S210) may be a step in which the string instrument value evaluation device analyzes the state information of the string instrument from the captured image.

The status information may include at least one of information about the type of damage occurring inside or outside the string instrument, the size of the damage, the location of the damage, the number of damage, and whether or not the part is replaced.

Old string instruments may have various internal or external damages due to the passage of time or continuous use. These damages are classified in various ways depending on the location or shape of the cause, and depending on the classification, they can have different effects on the value of the instrument. You can.

Information on the type of damage may refer to information for distinguishing whether the damage corresponds to a crack caused by a general crack or worm damage caused by an insect invasion. The size of the damage may refer to information about the horizontal or vertical length of the damage. The location of the breakage can mean information about which part of the string instrument the breakage occurred in. The number of damages may refer to information about how many damages occurred in the string instrument. Information about whether a part has been replaced may mean information about whether a specific part has been replaced with another part during the process of repairing or repairing a string instrument. Information on these types of damage or whether or not parts should be replaced can be managed separately in the resume of the relevant string instrument or updated through the process of analyzing captured images of the string instrument.

Through the above-described process, string instrument shooting images, such as internal images or external images in which noise is reduced or clarity is improved, can be provided to an image analysis model to derive state information. In another embodiment, the status information may be directly input into the string instrument value evaluation device by the user for each item through the user terminal.

More specifically, the captured image can be input to a deep learning model learned through neural network learning to derive information such as information on the type of damage, size of damage, and number of damages.

A representative example of the deep learning model may be CNN (Convolutional Neural Network), and this deep learning model may be learned through learning data in a separate device and then loaded into a string instrument value evaluation device and utilized. In addition, the string instrument valuation device provides captured images to an external computing device, the computing device conducts inferences through the captured images to derive information about the damages, and provides the derived information back to the string instrument valuation device. You can also provide it.

To describe CNN among the deep learning models in more detail, a computing device can use CNN (Convolutional Neural Networks) to perform neural network learning on the damage of a string instrument. Here, CNN may be a basic algorithm that can recognize patterns in damage data. A CNN consists of a series of neurons organized into layers, with each layer having learnable weights and biases.

A CNN's convolutional layer may consist of a kernel (or filter) and a feature map. The convolution operation is performed with a specified convolution kernel size, and the kernel can be operated as a convolution layer neuron with a learnable weight matrix. Strides can represent the number of steps the kernel moves across the pixel matrix of the input image, referring to the previous layer.

The output of the convolutional layer can be used by an activation function to generate a convolutional feature map. A convolutional feature map is the output of applying a filter to the previous layer, which can be passed as input to the next convolutional layer, and more complex features can be extracted by systematically stacking multiple convolutional layers.

The pooling layer reduces the parameters obtained from the convolution layer and improves computational efficiency, and the pooling layer can be classified into at least one of maximum, average, and random pooling. Here, average and maximum pooling determine the average or maximum value of adjacent neurons, respectively, while random pooling can select the value of a neuron according to a given probability. Additionally, the pooling layer can reduce the resolution of the convolutional feature map, thereby reducing the computational workload for network training.

Fully Connected Layers consist of multiple hidden layers, each of which may contain many neurons that are fully connected to neurons in subsequent layers.

The input layer can essentially be a one-dimensional vector obtained after performing convolution and pooling operations. A fully connected layer can map these extracted features to the final classification output. For example, Softmax and Support Vector Machines (SVM) can provide classification functions at the output layer of a fully connected layer.

Activation and loss functions are important building blocks of CNNs, and since the activation function is nonlinear, the network can be trained according to a nonlinear mapping. As an example, activation functions may include Rectified Linear Unit (ReLU), Leaky ReLU, tanh, and Sigmoid functions. A loss function or objective function can quantify the difference between predicted and actual values. Cross-entropy is a loss function used in multi-class classification, while mean square error can be frequently used in regression of continuous values.

Here, the computing device may include additional L1 and L2 regularization units in the loss function to mitigate overfitting that occurs in deep CNNs. L1 and L2 regularization are used to constrain the parameters of the loss function, and the main goal of backpropagation in deep CNN training is to minimize the loss function.

The computing device may perform transfer learning as an image evaluation model for evaluating damage to a string instrument using an image training model on which neural network learning was performed. Here, transfer learning can be an efficient approach used for learning on small data sets, applying a neural network pre-trained on a huge data set in the source domain to the target domain. Transfer learning allows common features learned in a sufficiently large data set to be transferred to another data set. Computing devices can perform transfer learning using feature extraction techniques and fine-tuning techniques.

The proposed image evaluation model can be trained on a data set containing all similar string damage damage images according to the background, as the background contains more or less noise outside the region of interest. Accordingly, the present invention minimizes cases of incorrect predictions and can successfully predict classes related to string instrument damage even though the tilt of the camera view and the luminance of the damage image vary.

The basic value derivation step (S220) may be a step in which the string instrument value evaluation device derives the basic value of the string instrument based on production information.

The production information of the string instrument may include at least one of the following: information about the region where the string instrument was produced, the manufacturer of the string instrument, and the year of production of the string instrument. For information on the regions where stringed instruments were made, refer to the regions where famous European violin makers were active, for example Cremona in Italy, Markneukirchen in Germany, and Mircourt in France. It may mean the information being displayed. Information about the maker of a stringed instrument may mean, for example, information about famous violin makers such as Antonio Stradivari, Giovanni Battista Guadagnini, or Martin Bowers. Information about the production period of stringed instruments may refer to information about the period when historically famous violins were produced, such as the 17th and 18th centuries. Information on the year of manufacture may be expressed as a rough approximation if relevant data is lacking.

The string instrument value evaluation device can derive the basic value of the string instrument through the production information of the string instrument, and this basic value can be used to calculate the discount value of the string instrument or the current value of the string instrument to which price change information is applied, which will be described later.

The basic value of a string instrument is loaded from a string instrument price database or a string instrument trading site using the production year, production area, and general market information according to the manufacturer of the string instrument traded on the market by a string instrument valuation device, or by operating a string instrument valuation system. It can be determined by matching production information to a price list set in advance by the entity.

The discount value derivation step (S230) may be a step in which the string instrument value evaluation device derives the discount value by applying a discount rate to the basic value based on the state information.

More specifically, the step of deriving the discount value (S230) is performed through the status information of the string instrument, when the front plate, back plate, or scroll of the string instrument is classified as having been replaced, or when the sound post of the string instrument is classified as having a crack. , if a crack is classified as occurring in the base bar of the string instrument, if worm damage occurs in the string instrument, if a crack occurs in the front, back, or side plate of the string instrument, and a discount rate is applied according to the classification to determine the discount value. can be derived.

Figures 3 and 4 are diagrams for illustrating and illustrating the configuration of a violin in relation to damage information among state information in the method for evaluating the value of a stringed instrument according to an embodiment of the present invention.

Referring to Figures 3 and 4, you can see the location and name of each part that makes up the violin among string instruments. Among these, the scroll 310 in Figure 3 is a decorative part located above the neck of the violin, and plays an important role as a design element of the violin. The scroll 310 is used when the violinist holds and plays the violin, and when the violinist plays the violin. This is a part made for convenient storage or movement.

The sound post 410 in Figure 4 is a small wooden pillar inside the body of the violin and is installed vertically between the front and back plates. The sound post 410 plays an important role in forming the sound of the violin. The sound post 410 serves to transmit vibration while maintaining pressure and balance between the front and back plates inside the body of the violin. Additionally, the position and angle of the sound post (410) also affects the tone and size of the violin.

The base bar 420 in Figure 4 is a small wooden pillar located inside the front plate 430 of the violin and is installed below the F hole. The base bar 420 plays an important role in forming the tone of the violin, transmits vibrations within the front plate 430, and maintains the balance of the front plate 430. The position and size of the base bar 420 have a great influence on the sound of the violin and can be adjusted.

The front plate 430 of FIG. 4 is located on the upper part of the violin body and is made of wood. The front plate 430 of the violin transmits all vibrations and sounds of the violin and plays an important role in determining the sound of the violin. Additionally, there is an F-hole inside the top plate, which plays a role in making the violin's sound clearer.

The back plate 440 of FIG. 4 is located at the bottom of the body of the violin and, like the front plate 430, is made of wood. The back plate 440 plays an important role in determining the sound of the violin. The side plate 450 of FIG. 4 is a part that surrounds the circumference of the body of the violin and is made of wood. The side plate 450 surrounds the body of the violin and serves to connect the front plate 430 and the back plate 440. It also plays a role in controlling the size, shape, weight, and balance of the violin.

If the string instrument valuation device classifies the front, back, or scroll of the string instrument as having been replaced through condition information, a discount rate can be applied considering that the main part of the instrument has been replaced. For example, a discount rate of 50% can be applied if the front or back panel of a violin or viola is replaced, 30% if the scroll is replaced, and 40% if the front or back panel of a cello or contrabass is replaced, a discount rate of 30% can be applied if the scroll is replaced. If replaced, a 25% discount rate can be applied.

The stringed instrument value evaluation device determines the discount rate based on status information by determining that a main crack has occurred when the sound post of the stringed instrument is classified as having a crack, when the stringed instrument's bass bar is classified as having a crack, and when worm damage has occurred to the stringed instrument. can be applied.

For example, the string instrument valuation device may apply a discount rate of 20% if the front sound post crack of a violin or viola is less than 3cm, and a discount rate of 25% if it is more than 3cm. Additionally, a discount rate of 10% can be applied if the crack on the front sound post of the cello or contrabass is less than 5cm, and 20% if the crack is more than 5cm.

The string instrument valuation device can apply a discount rate of 30% when the crack on the back sound post of a violin or viola is less than 3cm, and 35% when the crack is more than 3cm. In addition, a discount rate of 20% can be applied if the sound post crack on the back panel of the cello or contrabass is less than 5cm, and a discount rate of 25% can be applied if the crack is more than 5cm.

The stringed instrument valuation device can apply a discount rate of 10% if the crack in the bass bar of the front plate of the violin or viola is less than 3cm, and 15% if the crack is more than 3cm. In addition, a discount rate of 5% can be applied if the crack on the front base bar of the cello or contrabass is less than 5cm, and a discount rate of 10% can be applied if the crack is more than 5cm.

The string instrument valuation device can apply a discount rate of 25% if the number of worm damage to violins and violas is 1 to 3 and the sum of the damage lengths due to worm damage is less than 5cm, and a discount rate of 40% if the sum of the damage lengths is more than 5cm. And when the number of worm damage for cello and contrabass is 1 to 3, a discount rate of 20% can be applied when the sum of the damage length from worm damage is less than 10m, and a discount rate of 30% can be applied when the sum of the damage length from worm damage is more than 10cm. there is.

The string instrument valuation device can apply a discount rate when a crack occurs in at least one of the front, back, and side plates of the string instrument, considering that a general crack has occurred.

For example, the string instrument valuation device can apply a discount rate of 3% if the number of cracks formed on the front, back, and side panels of a violin or viola is 1 to 5, and the sum of the crack sizes is less than 5 cm, and 8% if it is more than 5 cm. there is. Additionally, if the number of cracks formed on the front, back, and side panels of a violin or viola is 6 or more and the total size of the cracks is less than 15 cm, a discount rate of 12% can be applied, and if the crack size is more than 15 cm, a discount rate of 15% can be applied.

In addition, the string instrument valuation device can apply a discount rate of 3% if the number of cracks formed on the front, back, and side panels of a cello or contrabass is 1 to 5 and the total crack size is less than 10 cm, and 5% if it is more than 10 cm. Additionally, if the number of cracks formed on the front, back, and side panels of a cello or contrabass is 6 or more and the total size of the cracks is less than 30cm, a discount rate of 8% can be applied, and if it is more than 30cm, a discount rate of 10% can be applied.

The visualization and display step (S240) may be a step in which the string instrument value evaluation device visualizes and displays at least one of the basic value and the discount value of the string instrument. A detailed description of this step will be provided later.

Figure 5 is a flowchart illustrating a method for evaluating the value of a string instrument according to an embodiment of the present invention.

Referring to Figure 5, the method for evaluating the value of a string instrument includes a current photographed image and production information reception step (S500), a current state information analysis step (S510), a basic value derivation step (S520), a state change information derivation step (S530), and a price. It may include a change information derivation step (S540), a present value derivation step (S550), and a visualization and display step (S560).

In the embodiment of FIG. 5, the current captured image and production information receiving step (S500), the current state information analysis step (S510), and the basic value derivation step (S520) are the same as the state information analysis step (S210) and basic value derivation in FIG. 2. Since it has already been explained through step S220, duplicate description of the same configuration will be omitted and the differences will be explained.

The currently captured image and production information receiving step (S500) may be a step in which the string instrument value evaluation device receives the currently captured image of the string instrument and production information of the string instrument.

The current state information analysis step (S510) may be a step in which the string instrument value evaluation device analyzes the current state information of the string instrument from the currently captured image. In another embodiment, the current state information may be input for each item directly into the string instrument value evaluation device by the user terminal rather than the currently captured image.

The basic value derivation step (S520) may be a step in which the string instrument value evaluation device derives the basic value of the string instrument based on the production information.

In the state change information derivation step (S530), the string instrument value evaluation device photographs the string instrument in the past and loads past state information of the string instrument analyzed from past captured images generated in the past, and converts the past state information into the current state information. This may be a step of deriving state change information by comparing with .

Here, the current captured image and the past captured image include an external image photographed by each part of the exterior of the string instrument and an internal image photographed by each region of the interior of the string instrument, and the internal image is photographed by at least one of X-rays, endoscopes, and ultraviolet rays. It could be a photo.

In addition, the internal images of the currently captured image and the past captured image may be images processed through at least one filter selected from the group consisting of a Gaussian filter, a median filter, a Laplacian filter, and a wavelet filter to remove noise from the captured image.

Additionally, the production information may include at least one of information about the region where the string instrument was manufactured, the manufacturer of the string instrument, and the year of production of the string instrument.

In addition, the past state information and current state information may include at least one of the following information about the type of damage that occurred inside or outside the string instrument, the size of the damage, the location of the damage, the number of damages, and whether or not the parts should be replaced.

The past state information may be state information created in advance by the string instrument value evaluation device from past captured images generated at a previous point in time. On the other hand, when only past captured images are stored and no past state information, the string instrument value evaluation device can extract state information from past captured images and use it as past state information.

In order to facilitate comparison of state information extracted from a past captured image and a current captured image, both the past captured image and the current captured image may be images shot of the same string instrument by the same photographing device using the same photographing method. For example, at a certain point in time in the past, the inside of a stringed instrument was changed clockwise by 60 degrees clockwise based on the f-hole of the stringed instrument using an endoscope to obtain six internal images, and each of the parts exposed to the outside was photographed. If 10 externally captured images were obtained, the current captured image also changes direction clockwise by 60 degrees based on the f-hole of the same string instrument to obtain 6 internally captured images, and each of the parts exposed to the outside is photographed to obtain 10 external images. It can be photographed in a way that obtains a photographed image. And for the same reason, a noise filtering algorithm can be applied to past and currently captured images to remove noise from images captured in the same way.

In the current state information analysis step (S510), the string instrument value evaluation device derives state information through an image analysis model for the noise-filtered captured image, similar to the method described in the state information analysis step (S210) in the embodiment of FIG. 2. You can. At this time, it is desirable that the string instrument value evaluation device applies the same image analysis model to both past and current captured images. Through this, the string instrument valuation device can derive the same class regarding string instrument damage for past and current images.

As described above, the damage class extracted for past and current images is at least one of the information about the type of damage that occurred inside or outside the string instrument, the size of the damage, the location of the damage, the number of damage, and whether or not the part was replaced. It may contain more than one.

The state change information derivation step (S530) determines whether to replace the front panel of the string instrument and replace the back panel during the period from the generation of the past state information to the generation of the current state information based on the production information, past state information, and the current state information. At least one of the information about whether or not the scroll is replaced, the rate of increase in the size of the front sound post crack, the rate of increase in the size of the back sound post crack, the rate of increase in the size of the base bar crack, the rate of increase in the size of worm damage, and the difference in the timing of status information generation can be derived.

If past state information and current state information are derived, the string instrument valuation device compares the two state information to determine whether the front, back, or scroll of the string instrument is replaced, the rate of increase in the size of the front sound post, the rate of increase in the size of the crack in the back sound post, and the size of the bass bar crack. It is possible to derive the increase rate, worm damage size increase rate, etc. In addition, the string instrument value evaluation device can compare the creation time of past state information and the creation time of current state information to derive how many days have passed since the past state information was created.

For example, in the past image taken, there was a 5 cm front sound post crack, and in the current image taken again after some time had passed, a 6 cm front sound post crack was formed. This means that the front sound post crack has an increase rate of 20%. can see.

In addition, the past state information and current state information may further include sound quality evaluation information, where the sound quality evaluation information is information evaluated through performance of the string instrument, and the sound quality evaluation information of the past state information generates the past state information. It may be sound quality evaluation information at the point in time. And the sound quality evaluation information of the current state information may be sound quality evaluation information at the time of generating the current state information. At this time, in order to accurately compare the sound quality evaluation information, the same string instrument is played in the same way, but a blind sound quality test is performed so that the timing of performance is not known, so that the sound quality evaluation information can be compared and quantified sound quality evaluation information can be generated. there is. Additionally, the change rate of the sound quality evaluation information can be calculated by comparing the sound quality evaluation information of the evaluated past state information with the sound quality evaluation information of the current state information.

That is, the string instrument value evaluation device may derive the state change information by calculating the change value of the sound quality evaluation information of the past state information and the current state information in the state change information deriving step (S530).

The price change information derivation step (S540) may be a step in which the string instrument valuation device derives price change information by applying the state change information to a regression model.

Here, the regression model may be a linear model for multiple regression analysis that sets at least one of the information derived in the state change information derivation step as an independent variable and uses the price change value as a dependent variable.

More specifically, the regression model may be a model calculated with the following formula.

[formula]

Y = a1*X1 + a2*X2 + a3*X3 + a4*X4 + a5*X5 + a6*X6+ a7*X7 + a8*X8 + a9*X9

Here, Y may mean the price change value included in the price change information generated by the string instrument valuation device. X1 is an independent variable with a value of 0 or 1 depending on whether the front or back plate of the string instrument is replaced, X2 is an independent variable with a value of 0 or 1 depending on whether the scroll replacement is replaced, and X3 is the sound post crack of the front plate. It is an independent variable with a value according to the size increase rate, X4 is an independent variable with a value according to the size increase rate of the sound post crack on the backboard, X7 is an independent variable with a value according to the rate of increase, is an independent variable having , and

And a1 to a9 may be coefficients for each independent variable set by the operator of the string instrument valuation system or the designer of the string instrument valuation method. The values of a1 to a9 can be calculated and set in advance in a way to minimize the error value, such as the mean square error for the regression model, based on the fitting of the dependent and independent variables that are past sample data, and the state change information to be described later. It can also be modified by saving and updating the regression model.

The current value derivation step (S550) may be a step in which the string instrument value evaluation device derives the current value of the string instrument by reflecting the price change information in the basic value.

When the price change value is derived by the above-mentioned regression model, the string instrument value evaluation device can derive the corrected current value of the string instrument by applying the price change value to the basic value. The current value of the string instrument derived in this way can be viewed as the value of the string instrument finally derived by the string instrument valuation device.

And after the step of deriving the current value (S550), the string instrument value evaluation device may store the derived current value and the state change information and update the regression model through the stored information.

The visualization and display step (S560) may be a step in which the string instrument value evaluation device visualizes and displays at least one of the basic value and the current value of the string instrument.

The specific operation process of the visualization and display step (S240) of FIG. 2 and the visualization and display step (S560) of FIG. 5 will be explained with reference to FIGS. 6 and 7.

Figures 6 and 7 are diagrams illustrating an example of a user interface screen that visualizes and displays a method for evaluating the value of a stringed instrument according to an embodiment of the present invention.

Figure 6 illustrates a user interface 600 for a user to input production information to a string instrument value evaluation device through a user terminal. Referring to FIG. 6, the user interface 600 for inputting production information displayed on the user terminal may include a production information input area 610, a status information input area 620, and a damage information input area 630. .

The production information input area 610 may be an area where a user inputs production information related to a string instrument. More specifically, the production information input area 610 may include an area for inputting information about the type of instrument, type of bow, manufacturer, instrument warranty, and production year of the string instrument.

The status information input area 620 may be an area where status information about a string instrument is input. The status information input area 620 is an area where the number of general cracks is input, and damage-related information about front sound post cracks, back sound post cracks, base bar cracks, and worm damage among main damage can be input. The input of such information can be analyzed from captured images of the inside or outside of the string instrument and automatically input by the string instrument value evaluation device, or the user can input it directly through the user terminal.

The damage information input area 630 is an area where the damaged part and damage type of the string instrument can be input in more detail in relation to the status information input area 620. Damage-related information can be entered for each of the high-pitched side panel, front, rear, and low-pitched side panel of the string instrument. The input of such information can be analyzed from captured images of the inside or outside of the string instrument and automatically input by the string instrument value evaluation device, or the user can input it directly through the user terminal.

FIG. 7 is a diagram illustrating a user interface 700 that displays the basic value, discounted value, or current value of a stringed instrument derived by a stringed instrument valuation device.

Referring to FIG. 7, the user interface 700 for displaying the value of a string instrument includes a production information display area 710, a status information display area 720, a damage information display area 730, and a value information display area 740. It can be included.

The production information display area 710 may be an area that displays previously input production information.

The state information display area 720 may be an area that displays state information used during the string instrument value evaluation process and the depreciation rate reflected in the value of the string instrument through the state information.

The damage information display area 730 may be an area that displays damage-related information among the condition information used during the string instrument value evaluation process, and also displays any damage information that has changed compared to past condition information.

The value information display area 740 may be an area that displays the basic value and discounted value or basic value and current value of the string instrument derived by the string instrument valuation device.

The string instrument valuation method may be further divided into additional steps or combined into fewer steps through the embodiments described above with reference to FIGS. 1 to 7. Additionally, some steps may be omitted or the order between steps may be switched as needed.

The above method of evaluating the value of a stringed instrument can also be implemented in the form of a computer program stored in a computer-readable recording medium executed by a computer or a recording medium containing instructions executable by a computer. Additionally, the above-described string instrument value evaluation method may also be implemented in the form of a computer program stored in a computer-readable recording medium that is executed by a computer.

Computer-readable recording media can be any available media that can be accessed by a computer and includes both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable recording media may include computer storage media. Computer storage media includes both volatile and non-volatile, 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.

The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as single may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims described below rather than the detailed description above, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.

1: Stringed instrument valuation system
100: Stringed instrument valuation device
200: Network
300: User terminal

Claims (29)

delete delete delete delete delete delete delete delete delete delete delete In the string instrument valuation method using a string instrument valuation device,
Receiving production information including a currently captured image of a string instrument and information on a region where the string instrument was produced;
analyzing current state information of the string instrument from the currently captured image;
Deriving a basic value of the string instrument based on the production information;
Loading past state information of the string instrument analyzed from a past photographed image generated in the past by photographing the string instrument in the past, and comparing the past state information with the current state information to derive state change information;
Deriving price change information by applying the state change information to a regression model;
deriving the current value of the string instrument by reflecting the price change information in the basic value; and
Visualizing and displaying at least one of the basic value and the current value of the string instrument; Including,
The current state information and the past state information include information about the type of damage of the string instrument,
The information on the type of damage is information for distinguishing whether the damage corresponds to a crack caused by a crack or worm damage caused by an insect intrusion,
The visualization and display step is a step of further displaying information related to damage among the status information,
The step of deriving the state change information is
Based on the production information, the past state information, and the current state information, whether the front plate, back plate, or scroll of the string instrument is replaced during the period from the time of generation of the past state information to the time of generation of the current state information, and whether the front plate sound post cracks Derive at least one of the size increase rate, back plate sound post crack size increase rate, base bar crack size increase rate, worm damage size increase rate, and information on the difference in status information generation time, and based on the derived current value and the status change information. update the regression model,
The regression model is
It is a linear model for multiple regression analysis in which at least one of the information derived in the derivation step of the state change information is set as an independent variable and the price change value is the dependent variable, and the regression model is independent. A method of evaluating the value of string instruments, characterized in that the coefficients for each variable are modified through updating the regression model. According to clause 12,
The basic value of the stringed instrument is determined by matching the production information according to the production region of the stringed instrument to a stringed instrument price database, a stringed instrument trading site, or a preset price list. According to clause 12,
The step of deriving the state change information includes a crack size increase rate of the string instrument during a period from the time of generation of the past state information to the time of generation of the current state information based on the production information, the past state information, and the current state information, and This is the step of deriving the increase rate of the worm damage size,
The step of deriving the price change information is a step of deriving the price change information by applying state change information including the crack size increase rate and the worm damage size increase rate to a shopping mall regression model. According to clause 12,
The information on the type of damage includes information on the size of the damage, such as how long the damage is horizontally or vertically, information on the number of damages, such as how many damages occurred in the string instrument, and information on the part of the string instrument in which the damage occurred. A method of evaluating the value of a stringed instrument, characterized in that it includes information on the location of damage. According to clause 15,
In the step of visualizing and displaying, the size information of the damage, the number of damages, and the location information of the damage of the string instrument are displayed in a status information display area, such as a discount rate,
Visualizing the damage size information, the number of damage information, and the location information of the damage and displaying them in a damage information display area, comparing the past state information and the current state information and displaying any changed damage information together. A method for evaluating the value of string instruments, characterized in that: According to clause 12,
The current captured image and the past captured image include an external image captured by each portion of the exterior of the string instrument and an internal image captured by each portion of the interior of the string instrument,
A method for evaluating the value of a string instrument, characterized in that the internal image is a photograph taken by at least one of X-rays, endoscopes, and ultraviolet rays. According to clause 17,
The internal image is an image processed through at least one of a Gaussian filter, a median filter, a Laplacian filter, and a wavelet filter to remove noise from the captured image. According to clause 12,
The production information includes at least one of information about the manufacturer of the string instrument and the year of production of the string instrument. According to clause 12,
A method for evaluating the value of a string instrument, wherein the past state information and current state information include information on whether or not parts occurring inside or outside the string instrument have been replaced. delete delete delete delete According to clause 12,
The regression model is calculated using the following formula,
<Formula>
Y = a1*X1 + a2*X2 + a3*X3 + a4*X4 + a5*X5 + a6*X6+ a7*X7 + a8*X8 + a9*X9
The Y is a price change value included in the price change information generated by the string instrument valuation device. The X1 is an independent variable with a value depending on whether the front or back plate of the string instrument is replaced, the X4 is an independent variable with a value according to the increase rate, X4 is an independent variable with a value according to the increase rate of the size of the sound post crack of the back plate of the string instrument, The X6 is an independent variable with a value according to the increase rate of the size of the worm damage of the string instrument, the It is an independent variable, and the
The a1 to a9 are coefficients for each independent variable set by the operator of the string instrument valuation system or the designer of the string instrument valuation method, and the values of a1 to a9 are determined by storing the state change information and updating the regression model. A method of valuing stringed instruments, characterized by modifications. According to clause 12,
The past state information and current state information further include sound quality evaluation information,
The sound quality evaluation information is information evaluated through the performance of the string instrument, the sound quality evaluation information of the past state information is sound quality evaluation information at the time of generating the past state information, and the sound quality evaluation information of the current state information is the sound quality evaluation information of the current state information generated. A string instrument value evaluation method characterized by sound quality evaluation information at the point of view. According to clause 26,
The step of deriving the state change information is
A method for evaluating the value of a stringed instrument, characterized in that the step of deriving the state change information by further calculating a change value in sound quality evaluation information of the past state information and the current state information. delete A computer program stored on a computer-readable recording medium comprising a sequence of instructions providing a method for valuing a stringed instrument, comprising:
Receive production information including a currently captured image of a string instrument and information about the region where the string instrument was manufactured,
Analyze current state information of the string instrument from the currently captured image,
Derive the basic value of the string instrument based on the production information,
Load past state information of the string instrument analyzed from past captured images generated in the past by photographing the string instrument in the past, and compare the past state information with the current state information to derive state change information, wherein the production information, Based on the past state information and the current state information, whether the front plate, back plate, or scroll of the string instrument is replaced during the period from the time of generation of the past state information to the time of generation of the current state information, the rate of increase in the size of the front sound post crack, and the back plate Derive at least one of information about the sound post crack size increase rate, base bar crack size increase rate, worm damage size increase rate, and difference in status information generation time, and update the regression model based on the derived current value and the state change information. ,
Applying the state change information to the regression model to derive price change information,
Deriving the current value of the string instrument by reflecting the price change information in the basic value,
Visualizing and displaying at least one of the basic value and the current value of the string instrument,
The current state information and the past state information include information about the type of damage of the string instrument,
The information on the type of damage is information for distinguishing whether the damage corresponds to a crack caused by a crack or worm damage caused by an insect intrusion,
Among the above status information, more information related to damage is displayed,
The regression model is
It is a linear model for multiple regression analysis in which at least one of the information derived in the derivation step of the state change information is set as an independent variable and the price change value is the dependent variable, and the regression model is independent. A computer program stored in a computer-readable recording medium, including a sequence of instructions in which coefficients for each variable are modified through updating the regression model.

Images