24-hour unmanned automatic car wash system with remote control

Abstract

The present invention relates to a remotely controllable 24-hour unmanned automatic car wash system that integrates member information, vehicle registration information, usage ticket information, car wash operation status, and payment information.

Description

Technology Field

[0001] The present invention relates to a remotely controllable 24-hour unmanned automatic car wash system, and more specifically, to a remotely controllable 24-hour unmanned automatic car wash system that integrates and manages member information, vehicle registration information, usage ticket information, car wash operation status, and payment information. Background Technology

[0003] With the widespread adoption of vehicles, self-service car washing has recently established itself as a hobby. Traditionally, payment methods at self-service car washes have primarily relied on cash. In other words, users exchange banknotes for coins at a coin exchange machine and then use the equipment for a set amount of time by inserting either the coins or the banknotes directly into the machine.

[0004] The method of using cash in this way was very inconvenient for users as they had to carry cash, and the management process was very complex for managers as they had to collect the cash.

[0005] Although a method of pre-charging fees onto prepaid cards is being used to replace cash payments, there are instances where payment errors cannot be resolved quickly because administrators are unable to monitor the situation in real time.

[0006] In addition, if a user wanted a refund for a prepaid card, the management process was complicated because the administrator had to personally check the balance and process the refund.

[0007] Furthermore, when self-service car washes are established using cash or prepaid card methods, usage statistics for each machine are not automatically aggregated; therefore, usage was estimated indirectly by separately calculating the revenue for each machine. Consequently, it is difficult to manage the machines efficiently as the hourly usage of each machine cannot be accurately determined.

[0008] Meanwhile, 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 necessarily be considered publicly known technology disclosed to the general public prior to the filing of the present invention. Prior art literature

[0010] Korean Published Patent No. 10-2017-0017234 The problem to be solved

[0011] One aspect of the present invention provides, in detail, a remotely controllable 24-hour unmanned automatic car wash system that integrates member information, vehicle registration information, usage ticket information, car wash operation status, and payment information.

[0012] The technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0014] A remotely controllable 24-hour unmanned automatic car wash system according to one embodiment of the present invention includes a car wash management server that communicates with a user terminal to provide unmanned car wash-related services.

[0015] The above car wash management server is,

[0016] A pass management unit that communicates with a user terminal possessed by a user using a car wash to provide membership registration and payment services;

[0017] A remote car wash management unit that communicates with a kiosk installed at a car wash and controls the kiosk through the user terminal; and

[0018] Includes the sales status management department.

[0019] The aforementioned pass management department,

[0020] A list of car washes registered in the above car wash management server is sorted and displayed according to one of the criteria of distance, name, newest business, or business rating, and a payment process for a usage pass for at least one car wash selected by the user terminal is performed.

[0021] The above usage right is,

[0022] It is characterized by being one of the following: a general pass that allows one to use a car wash selected from the user terminal once; a general subscription pass that allows one to use a car wash selected from the user terminal within the range of the prepaid amount by prepaying a predetermined amount; and an unlimited subscription pass that allows one to use the car wash without limit upon payment of a predetermined amount.

[0023] The aforementioned pass management department,

[0024] A user index is calculated based on consumer information received from the user terminal and usage voucher information paid for by the user terminal, and a car wash tailored to the consumer is recommended based on the calculated user index.

[0025] The above car wash management server is,

[0026] It further includes a car wash guide unit that scans the contamination level of a vehicle intending to use the car wash and transmits the scanning result to a user terminal, and

[0027] The above car wash guide section is,

[0028] A captured image of a vehicle is analyzed to divide the captured image into predetermined pixel units, brightness information, texture information, and edge information of each divided pixel block are extracted, and the extracted brightness information, texture information, and edge information are input into a pre-trained artificial neural network to calculate a contamination score for each pixel block based on the output value of the artificial neural network.

[0029] The above car wash guide section is,

[0030] A pollution map is generated in which pixel areas where the pollution score output from the artificial neural network is greater than or equal to a preset threshold value are represented in the first color, and pixel areas where the pollution score output from the artificial neural network is less than the preset threshold value are represented in the second color, and the generated pollution map is transmitted to a user terminal. Effects of the invention

[0032] According to one aspect of the present invention described above, member information, vehicle registration information, usage ticket information, car wash operation status, and payment information can be managed in an integrated manner, thereby improving the convenience of use for consumers and the convenience of car wash management for car wash owners.

[0033] In addition, it can reliably recommend customized car washes based on user information. Brief explanation of the drawing

[0035] FIG. 1 is a diagram showing the schematic configuration of a remotely controllable 24-hour unmanned automatic car wash system according to one embodiment of the present invention. Specific details for implementing the invention

[0036] The following detailed description of the invention refers to the accompanying drawings, which illustrate specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that various embodiments of the invention are different but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be implemented in other embodiments without departing from the spirit and scope of the invention in relation to one embodiment. It should also be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the invention. Accordingly, the following detailed description is not intended to be limiting, and the scope of the invention is limited only by the appended claims, including all equivalents to those claimed therein, provided appropriately described. Similar reference numerals in the drawings refer to the same or similar functions across various aspects.

[0037] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the drawings.

[0038] FIGS. 1 and 2 are drawings illustrating the schematic configuration of a remotely controllable 24-hour unmanned automatic car wash system according to an embodiment of the present invention.

[0039] The remotely controllable 24-hour unmanned automatic car wash system according to the present invention aims to comprehensively manage member information, vehicle registration information, usage ticket information, car wash operation status, and payment information.

[0040] To this end, a remotely controllable 24-hour unmanned automatic car wash system includes a car wash management server.

[0041] The car wash management server communicates with user terminals to provide unmanned car wash-related services to users utilizing the car wash facility. Through this, users can conveniently access various services, such as identifying nearby car washes, purchasing usage passes, and controlling car wash equipment during washing, via software (applications) installed on their terminals.

[0042] Such a car wash management server can build a database containing data collected from users. The learning management server may include a program that automatically builds a database using open source software.

[0043] The car wash management server may be a self-contained server, a cloud server, or a peer-to-peer (P2P) set of distributed nodes. The facility database construction server (100) may include a processor and memory. The processor may execute a program or control the facility learning management server. Program code executed by the processor may be stored in memory. The memory may store information related to the present invention or a program for implementing a method. The memory may be volatile memory or non-volatile memory.

[0044] The car wash management server can be connected to a user terminal via a network. Networks include, but are not limited to, the Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), 3G, 4G, LTE, 5G, Wi-Fi, etc.

[0045] In this way, the remotely controllable 24-hour unmanned automatic car wash system according to the present invention can provide customers with various payment options and maximize the convenience of using the car wash service.

[0046] In addition, by supporting a waiting payment function, subsequent vehicles can pay for the car wash in advance even if there is a vehicle ahead, thereby reducing waiting time and improving car wash efficiency.

[0047] In addition, car wash owners can conveniently view sales status, including revenue inquiries and member management, through the administrator page. Remote control is possible even during unmanned operation, enabling efficient management of car wash machines. Furthermore, system administrators can remotely monitor and control car wash machines and kiosks, supporting more effective and stable operations.

[0048] Specifically, the car wash management server includes a usage right management department, a remote car wash management department, and a sales status management department.

[0049] The pass management department communicates with the user terminal carried by the user using the car wash to provide membership registration and payment services.

[0050] In one embodiment, the pass management unit sorts and displays a list of car washes registered in the car wash management server according to one of the criteria of distance, name, newest business, or business rating, and performs a pass payment process for using at least one car wash selected by the user terminal.

[0051] At this time, the usage right may be in the form of any one of the following: a general usage right that allows one to use a car wash selected from the user terminal once; a general subscription right that allows one to use a car wash selected from the user terminal within the range of the prepaid amount by prepaying a predetermined amount; and an unlimited subscription right that allows one to use the car wash without limit upon payment of a predetermined amount.

[0052] The pass management department ensures that the user terminal screen displays General Pass, General Subscription, and Unlimited Subscription according to the type of ticket sold, and if tickets of each type (pass) are unavailable at the selected car wash, a message such as "Service usage is restricted because the available quantity of the currently purchasable ticket type car wash pass has been exceeded" is displayed at the car wash name.

[0053] In addition, the pass management department ensures that when a subscription is restricted according to the subscriber limit set for each branch, the waiting list and the number of waiting customers for each branch attempting to purchase are displayed.

[0054] In other words, when the subscriber limit is reached for a car wash branch, it enters a waiting state, and clicking the waiting notice displays a waiting application form. Once the waiting application is completed, the current waiting order (number) is displayed, and when the status changes from waiting to available for purchase due to the user canceling their subscription or the administrator adjusting the subscription limit, the user is notified.

[0055] The remote car wash management unit communicates with a kiosk installed at the car wash and controls the kiosk through the user terminal.

[0056] Through this, when a user selects one of the different car wash courses displayed on the user terminal, the remote car wash management unit transmits a control signal to the kiosk to enable car washing for the selected course to be performed.

[0057] The remote car wash management department ensures that the database regarding the user's car wash course selection is stored on the car wash management server.

[0058] The sales status management department communicates with the owner's terminal held by the car wash owner to transmit information such as the car wash pass purchase status, sales status, pass cancellation status, and car wash rating to the owner's terminal.

[0059] In some other embodiments, the pass management unit calculates a user index based on consumer information received from a user terminal and pass information paid for by the user terminal, and recommends a car wash tailored to the consumer based on the calculated user index.

[0060] Previously, car washes were recommended using only fragmentary information, such as ratings based on review data or locations near the consumer's residence, which resulted in a problem where only uniform recommendation services were provided that failed to reflect consumer characteristics.

[0061] To solve these problems, the usage right management unit according to the present invention is characterized by recommending a car wash tailored to the consumer by considering information on the consumer's actual car wash usage patterns.

[0062] To this end, the pass management department sets a first setting value based on the gender of the consumer (car wash user), a second setting value based on the age group of the consumer, the average value of the distance between the user's residence and the car wash used, the average usage time per car wash use, the payment amount for a general pass, the payment amount for a general subscription pass, and the payment amount for an unlimited subscription pass as variables, and calculates a user index based on the set variables.

[0063] Here, the first setting value, which is set according to the consumer's gender, can be set to a value of 1 if the consumer is male and a value of 10 if the consumer is female, and this setting value is intended to distinguish the propensity score calculated according to the consumer's propensity, because a higher or lower user index is not necessarily better or worse.

[0064] The second setting value, which is set according to the age group of the consumer, can be set to a value of 1 if the consumer is in their teens, 2 if they are in their 20s, 3 if they are in their 30s, 4 if they are in their 40s, 5 if they are in their 50s, and 6 if they are 60 or older. This setting value is intended to distinguish the propensity score calculated according to the age group of the consumer, because a higher or lower user index is not necessarily better or worse.

[0065] The average usage time per car wash visit can be set to a value of 5 if less than 30 minutes, 10 if 30 minutes or more but less than 1 hour, 15 if 1 hour or more but less than 3 hours, and 20 if 3 hours or more.

[0066] The average distance between the user's residence and the car wash used can be set to a value of 3, for example, when the user's history of using the first car wash and the second car wash is found, and the distance between the residence and the first car wash is 2 km and the distance between the residence and the second car wash is 4 km.

[0067] A user index can be calculated by taking these variables into account, and specifically, it can be calculated in proportion to the ratio of the average usage time per car wash visit to the average distance between the user's residence and the car wash used, and in proportion to the ratio of the total payment amount of the general subscription and the unlimited subscription to the total payment amount of the general pass.

[0068] In particular, in the present invention, the reliability of the calculated index and the car wash recommendation using it can be improved by calculating a user index by considering factors such as whether the amount paid to use the car wash is a one-time payment (payment amount of a general pass), whether the car wash is visited multiple times (payment amount of a general subscription pass), and how often the car wash is visited (payment amount of an unlimited subscription pass, average usage time per car wash visit).

[0069] Meanwhile, in the aforementioned user index calculation process, each variable can be calculated by considering only its magnitude value. That is, each variable is considered only for its magnitude value, and the user index can be set using the magnitude value of the result calculated in this way.

[0070] In the process of calculating the user index described above, a logarithmic function may be applied to a specific operation value. The reason for applying the logarithmic function is that it can express large values ​​by relatively reducing them. That is, as the x value increases, the logarithmic function changes in a form that is less affected by units or scales. This allows the result value to react less sensitively to changes in units, enabling data of various scales to be compared on the same scale. Additionally, the logarithmic function expresses the change in x value as a linear value when the change in x value has an exponential difference. This allows the range of change in large values ​​to be appropriately reduced and represented, making it useful when the change in x value has a proportional nature. Furthermore, the logarithmic function may be applied to the operation value to prevent it from changing significantly only due to certain variable values ​​or operation values ​​according to the present invention.

[0071] In this way, the voucher management department calculates a user index. By comparing the user index calculated through the aforementioned process with the matching index pre-assigned to each car wash business in a stored database, the department can set the top N car washes with matching indices that have the most similar magnitude to the user index as car washes to be recommended to the user. That is, to recommend customized car washes to consumers, the voucher management department calculates the user index using various user information (age, gender, average car wash usage time, distance from the car wash, amount spent, etc.) as variables.

[0072] In some other embodiments, the car wash management server according to the present invention may further include a car wash guide section.

[0073] Traditional self-service car washes rely on users to visually inspect their vehicles and decide whether to wash and the intensity based on their subjective judgment. However, this approach carries the risk of inadequate cleaning due to factors such as poor visibility of contaminated areas, dim lighting, weather conditions, and viewing angles. Additionally, users struggle to determine which of the various washing tools—such as high-pressure water, foam guns, and brushes—to use.

[0074] To solve these problems, the car wash guide unit according to the present invention scans the contamination level of a vehicle intending to use a car wash and transmits the scanning result to a user terminal, thereby helping the user make a car wash decision more quantitatively and intuitively.

[0075] To this end, the car wash guide unit first collects video footage of a vehicle requiring washing. For example, the car wash guide unit can collect video footage from fixed or rotating cameras installed inside or at the entrance of a self-service car wash, preprocesses the vehicle exterior images input from the image acquisition unit, and segments the vehicle's exterior to estimate the degree of contamination in each area.

[0076] Specifically, the car wash guide unit corrects image distortion caused by factors such as lens distortion and corrects brightness and contrast, and then extracts vehicle objects from the captured image. To this end, the car wash guide unit divides the captured image into predetermined pixel units (e.g., 3*3 pixels) and extracts features of each divided pixel block. For example, the car wash guide unit extracts the average brightness, texture information, and edge information of the pixel blocks.

[0077] Texture information represents visual patterns indicating how regularly, repeatedly, or roughly pixel values ​​change within an image; when the vehicle surface is clean, the pixels constituting a pixel block will have uniform brightness and color. Conversely, in areas contaminated with dust or foreign matter, it is predicted that specific pixels within a pixel block will show distinct differences in brightness or color compared to others.

[0078] To this end, the car wash guide unit generates a Local Binary Pattern (LBP) by comparing the texture pattern with neighboring pixel blocks (3x3) surrounding the pixel block and expressing it as a binary number, whereby if the surrounding pixel block is brighter than the reference pixel block, it is 1, and if it is darker, it is 0.

[0079] Subsequently, the car wash guide unit detects pixels with various scales and orientations using a Gabor filter.

[0080] A Gabor filter is a selective frequency filter that responds to a given direction and frequency, making it suitable for detecting directional textures (e.g., oil stains, tire tracks, etc.) within an image. Through the output of the Gabor filter, it is possible to detect traces of contamination, such as directional stains, which are phenomena where continuous or repetitive patterns or linear textures appear in a specific direction within an image. For example, if mud stains are present on the side of a vehicle, a striped pattern may be formed as water splashed from the wheels flows in a specific direction; contaminants generated by insects colliding with the vehicle surface may form directional burst patterns, such as radial or spray patterns, on the front surface. The car wash guide can detect these directional patterns using a Gabor filter.

[0081] The present invention applies Gabor filters of different angles to detect directional contamination patterns in vehicle exterior images and determines that when response intensity is concentrated in a specific direction, it is a directional stain. Such characteristics frequently appear in areas such as flying debris from tires and mud runoff caused by vehicle movement, and are difficult to quantitatively determine using conventional image analysis techniques. To solve this problem, the present invention utilizes a Gabor filter to detect kernels vibrating in specific directions (e.g., 0°, 45°, 90°, etc.) to detect traces of contamination, such as directional stains.

[0082] For example, if there is a dirt stain extending at a 45° angle on the side of the vehicle, Gabor ( The response value at =45° becomes larger, while the response value at other angles is derived to be smaller. Therefore, the car wash guide can estimate the directionality of the detected stain.

[0083] More specifically, the car wash guide classifies vehicle stains in the image into one of side mud stains, stains formed as water dries after washing, insect impact marks, or road mud after rain or snow, based on the length-to-width ratio (L / w) of the stain, maximum Gabor response direction, vehicle exterior reference position, number of stain clusters, and shape characteristics based on Gabor response.

[0084] For example, if the stain formation location is the lower side of the vehicle, the angle at which the response value of the Gabor filter is highest is 45° or 135°, the L / W ratio is 3 or higher, and the pattern is continuous and the directionality is the same, the stain is classified as a side mud stain.

[0085] In addition, if the stain is formed on the side or rear of the vehicle, the angle at which the response value of the Gabor filter is highest is 85° to 105°, the L / W ratio is 1 to 2, and the stain is transparent, the stain is classified as a mark formed as water dries after washing the car.

[0086] In addition, if the stain formation location is the upper front part of the vehicle, there are 5 or more high angles of response value of the Gabor filter, and each stain is a diffuse type, the stain is classified as an insect impact mark.

[0087] Finally, if the stain formation location is around the vehicle wheel, the angle at which the response value of the Gabor filter is highest is 170° to 190°, the L / W ratio is 1 to 3, and the area of ​​the stain is 15% or more of the total area of ​​the vehicle, the stain is classified as road mud after rain or snow.

[0088] As such, the present invention can provide a more detailed car wash guide to the user by analyzing a directional contamination pattern formed on the exterior of a vehicle based on image processing indicators such as vehicle location information, directional texture response (Gabor filter), stain shape indicator (L / W ratio), and the number of stain clusters, and by automatically classifying the cause of the stain generation. In particular, it can automatically identify various types of contamination, such as mud, water stains, insect collision marks, and road mud marks, through a combination of the directionality and spatial distribution of the stain.

[0089] Next, the car wash guide unit extracts edge information from the image. Edge information refers to boundaries within the image where abrupt changes in brightness or color occur, and typically corresponds to the outlines of objects or contamination boundaries. To this end, the car wash guide unit removes noise from the image and then extracts the outlines of areas with significant brightness changes.

[0090] Specifically, the car wash guide detects edge candidates using the Sobel and Canny edge detection methods, and then detects the final edge using the method from which the clearer edge is extracted.

[0091] Canny edge detection is a method that removes noise from an image, detects the contours of areas with large brightness changes, and converts them into a binary image, while the edge detection method using a Sobel filter measures edge intensity by calculating derivatives (gradients) in the horizontal and vertical directions, and generates an edge intensity map by calculating the magnitude and direction of the gradient at each pixel.

[0092] Afterwards, the car wash guide section removes parts where the edge strength is not maximum to make the edge lines thin. Then, a threshold value is set according to the edge strength to distinguish between strong edges and weak edges, and edge information is extracted from the image by allowing weak edges to be connected.

[0093] In this way, the car wash guide unit extracts the average brightness, texture information, and edge information of each pixel block.

[0094] Subsequently, the car wash guide unit calculates a contamination score based on brightness information of the extracted pixel block, normalized Gabor response values, and edge pixel ratios. That is, the car wash guide unit can calculate the contamination score by summing the brightness information regarding the difference between the average brightness value of the pixel block and the overall brightness value of the image, the direction (angle) value having the highest response value among the Gabor filter response values ​​for each direction of the pixel block, and the ratio of the number of pixels detected as edges among all pixels constituting the pixel block, and dividing the sum by 3.

[0095] Finally, the car wash guide unit generates a pollution map in which pixel areas where the pollution score output from the artificial neural network described above is greater than or equal to a preset threshold value are represented as a first color, and pixel areas where the pollution score output from the artificial neural network is less than or equal to a preset threshold value are represented as a second color.

[0096] For example, the car wash guide unit converts pixels corresponding to vehicle objects in the captured image to white, and pixels corresponding to background areas excluding the vehicle to black. Subsequently, the car wash guide unit calculates the contamination level for pixels corresponding to vehicle objects, keeping pixel areas with a contamination score of less than 0.5 output from the artificial neural network white, and converting pixel areas with a contamination score of 0.5 or higher output from the artificial neural network to yellow or red, thereby making it easy to identify the location and area where stains are formed.

[0097] To describe an embodiment using this configuration of the present invention, when a user's vehicle enters the entrance of a car wash, an upper fixed camera photographs the vehicle. The system automatically extracts the vehicle area and generates a pollution map by analyzing the pollution level of each block using a CNN-based AI model. The pollution map is visualized as a heatmap and can be expressed as quantified information, such as an average pollution level of 0.68 and a high pollution ratio of 22%. Accordingly, the user can receive visualized information regarding the contaminated areas through a smartphone screen.

[0098] At this time, the present invention automatically classifies a car wash method by quantifying the average contamination intensity, the distribution ratio of high-contamination areas, directional texture response values, and edge density from a contamination map generated as a result of contamination analysis. For example, if a Gabor response in a specific direction is detected above a threshold value and the edge density is high, it is determined that physical contact cleaning is required and the use of a brush is recommended; if the overall contamination distribution is wide and the intensity is high, a dispersed cleaning method using a foam gun is recommended.

[0099] To this end, the system according to the present invention can construct a neural network that extracts contextual information for input data by learning training data with the Word2Vec algorithm to understand or estimate the meaning of the data.

[0100] The Word2Vec algorithm may include a Neural Network Language Model (NNLM). A Neural Network Language Model is fundamentally a neural network composed of an Input Layer, a Projection Layer, a Hidden Layer, and an Output Layer. The Neural Network Language Model is used as a method for vectorizing words. Since the Neural Network Language Model is a well-known technology, a more detailed explanation will be omitted.

[0101] The Word2vec algorithm is designed for text mining and determines proximity based on the preceding and succeeding relationships between words. It is an unsupervised learning algorithm. As its name suggests, Word2vec is a quantitative technique that represents the meaning of words in vector form. The Word2vec algorithm can represent each word as a vector in a space of approximately 200 dimensions. By utilizing the Word2vec algorithm, a vector corresponding to each word can be obtained.

[0102] The Word2vec algorithm can enable a dramatic improvement in precision in the field of natural language processing compared to other conventional algorithms. Word2vec learns the meaning of words by utilizing the relationships between words and adjacent words within sentences of an input corpus. Based on artificial neural networks, the Word2vec algorithm starts from the premise that words with the same context carry similar meanings. The algorithm learns through text documents, training the neural network to identify related words by identifying other words that appear nearby (approximately 5 to 10 words before and after) a given word. Since words with related meanings are highly likely to appear close together within a document, the two words can gradually acquire closer vectors as the learning process is repeated.

[0103] There are two training methods for the Word2vec algorithm: CBOW (Continuous Bag Of Words) and skip-gram. The CBOW method predicts a target word by utilizing the context created by surrounding words. The skip-gram method predicts words that may follow a single word. The skip-gram method is known to be more accurate in large-scale datasets.

[0104] Accordingly, in the embodiments of the present invention, a Word2vec algorithm using the skip-gram method is used. For example, if training is successfully completed through the Word2vec algorithm, similar words can be located nearby in a high-dimensional space. According to the Word2vec algorithm described above, the closer the distribution of surrounding words within a training document, the more similar the calculated vector values ​​can be, and words with similar calculated vector values ​​can be considered similar. Since the Word2vec algorithm is a known technology, a more detailed explanation regarding the calculation of vector values ​​will be omitted.

[0105] The server can input collected data into a neural network to extract an evaluation result vector value representing contextual information.

[0106] The server calculates the similarity between the evaluation result vector value and each of the multiple reference vector values, and can extract the reference vector value among the multiple reference vector values ​​that has the highest similarity to the evaluation result vector value. In this case, Euclidean distance, cosine similarity, Tanimoto coefficient, etc., may be adopted as the similarity calculation method.

[0107] The learning management server can extract the word corresponding to the reference vector value with the highest similarity to the evaluation result vector value as the word corresponding to the recognized text.

[0108] Furthermore, the learning management server can train artificial neural networks and utilize artificial neural networks that have completed training. The processor can train or execute artificial neural networks stored in memory, and memory can store artificial neural networks that have completed training. The electronic device that trains the artificial neural network and the electronic device that utilizes it may be identical, but they may also be separate. Artificial intelligence is a computer system that partially implements the functions of the human brain and is capable of learning, speculating, and making judgments on its own. As learning progresses, the probability of extracting the correct answer may increase. Artificial intelligence can be composed of learning and elemental technologies that utilize it. The learning of artificial intelligence is an algorithmic technology that classifies and learns features based on input data, and the elemental technologies may be technologies that partially implement the functions of the human brain by utilizing learning algorithms.

[0109] Artificial intelligence is a technology that facilitates the approach to problems where multiple probabilistic answers are possible, enabling it to logically and probabilistically infer optimal cycles, methods, and plans based on input data. AI inference techniques can include evaluating input data, optimization prediction, knowledge and probability-based reasoning, and preference-based planning.

[0110] Artificial neural networks are learning algorithms in the field of machine learning that programmatically implement the connections between neurons and synapses in the brain. By creating a neural network structure through programming and then training it, artificial neural networks can acquire desired functions. Although errors may exist, they can learn from massive datasets to produce appropriate output data from input data. They have the advantage of being able to obtain output data that has yielded statistically good results and are similar to human reasoning.

[0111] The server can construct query / metric datasets required for learning using artificial intelligence algorithms built on big data, and to this end, it may include multiple pre-trained artificial neural networks.

[0112] The system according to the present invention may include a plurality of pre-trained artificial neural networks for performing machine learning algorithms. Through machine learning, it can output output data based on input data and learn autonomously using the results, thereby improving its data processing capabilities. The artificial neural network can extract features and predict regularities based on input data to output result data, and as this process accumulates, the reliability of the result data increases.

[0113] In this embodiment, the artificial neural network may be an algorithm that outputs text data from at least one feature data among the shape, length, number, and height difference of an object recognized as text. The artificial neural network can infer the best output data by using big data as input data as is, or by using it as input data after undergoing a processing step to clean up unnecessary data.

[0114] Artificial intelligence machine learning models are classified according to the type of learning into Supervisory Learning, Unsupervisory Learning, Semi-supervised Learning, and Reinforcement Learning. In addition, machine learning algorithms include Decision Trees, K-Nearest Neighborhoods, Artificial Neural Networks, Support Vector Machines, Ensemble Learning, Gradient Descent, and Na Bayes Classifier, Hidden Markov Model, K-Means Clustering, etc. can be used.

[0115] Artificial neural networks can be pre-trained on various input values ​​that may be included in the input data. An artificial neural network can be trained using reinforcement learning, a learning method. Reinforcement learning is a method that gradually increases the probability of obtaining the correct result by setting rewards and constraints. Artificial neural networks can also be modeled based on Convectional Neural Networks (CNNs) or Recurrent Neural Networks (RNNs).

[0116] In this way, the system according to the present invention can estimate the meaning of text data using big data and artificial neural networks.

[0117] The technology according to the present invention, as described above, may be implemented in the form of program instructions that can be executed through various computer components or implemented as an application, and may be recorded on a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc., either individually or in combination.

[0118] The program instructions recorded on the above-mentioned computer-readable recording medium are those specifically designed and configured for the present invention, but may also be those known and available to those skilled in the art of computer software.

[0119] Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, and flash memory.

[0120] Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware device may be configured to operate as one or more software modules to perform processing according to the present invention, and vice versa.

[0121] Although the invention has been described above with reference to embodiments, those skilled in the art will understand that various modifications and changes can be made to the invention without departing from the spirit and scope of the invention as set forth in the following claims.

Claims

Claim 1 A remotely controllable 24-hour unmanned automatic car wash system comprising a car wash management server that communicates with a user terminal to provide services related to unmanned car wash, wherein the car wash management server comprises: a pass management unit that communicates with a user terminal possessed by a user using the car wash to provide membership registration and payment services; and a remote car wash management unit that communicates with a kiosk installed at the car wash to control the kiosk through the user terminal. The system includes a sales status management unit, wherein the usage right management unit sorts and displays a list of car washes registered in the car wash management server according to any one of the criteria of distance, name, newest business, or business rating, and performs a usage right payment process for using at least one car wash selected by the user terminal, wherein the usage right is characterized as being one of a general usage right that allows one to use the car wash selected by the user terminal once, a general subscription right that allows one to use the car wash selected by the user terminal within the range of the prepaid amount after prepaying a predetermined amount, and an unlimited subscription right that allows one to use the car wash without limit upon payment of a predetermined amount, and wherein the usage right management unit sets a first setting value set according to the consumer's gender, a second setting value set according to the consumer's age group, the average value of the distance between the user's residence and the car wash used, the average usage time per car wash use, the payment amount of the general usage right, the payment amount of the general subscription right, and the payment amount of the unlimited subscription right as variables, and calculates a user index based on the set variables. A remotely controllable 24-hour unmanned automatic car wash system characterized by recommending a consumer-customized car wash based on a calculated user index, wherein the user index is calculated in proportion to the ratio of the average usage time per car wash visit to the average value of the distance between the user's residence and the car wash used, and in proportion to the ratio of the sum of the payment amounts for a standard subscription and an unlimited subscription to the payment amount for a standard pass. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 In claim 1, the car wash management server further includes a car wash guide unit that scans the contamination level of a vehicle intending to use the car wash and transmits the scanning result to a user terminal, and the car wash guide unit analyzes a captured image of a vehicle to divide the captured image into predetermined pixel units, extracts brightness information, texture information, and edge information of each divided pixel block, inputs the extracted brightness information, texture information, and edge information into a pre-trained artificial neural network, and calculates a contamination score for each pixel block based on the output value of the artificial neural network, thereby enabling remote control of a 24-hour unmanned automatic car wash system. Claim 6 In claim 5, the car wash guide unit generates a pollution map in which pixel areas where the pollution score output from an artificial neural network is greater than or equal to a preset reference value are represented in a first color, and pixel areas where the pollution score output from the artificial neural network is less than or equal to a preset reference value are represented in a second color, and transmits the generated pollution map to a user terminal, thereby enabling remote control of a 24-hour unmanned automatic car wash system.

Images