KR101558486B1 - Method for providing tire information using a portable terminal - Google Patents

Abstract

The present invention relates to a tire information providing method using a portable terminal. According to an embodiment of the present invention, the tire information providing method using a portable terminal includes: a step of receiving a selection signal for an image search among search menus of a tire information providing system; a step of operating a camera module of the portable terminal in response to the received selection signal; a step of obtaining a tire image by photographing a predetermined tire; a step of analyzing the obtained tire image; and a step of providing the tire information according to the analysis result.

Description

TECHNICAL FIELD [0001] The present invention relates to a tire information providing method using a portable terminal,

The present disclosure relates to a tire information providing method using a portable terminal.

In the past, drivers' tires were purchased offline by purchasing goods only by the intention of the seller to search for the tire store, so that the tire could not know the exact information and could not compare the price.

As the online market has grown, online tire purchases have also increased and online tire sales sites have also been activated. These online purchases can be made at affordable prices by comparing prices.

Conventionally, an online product search method is a product search method using a vehicle type search and a size search, and is a product search method on the premise that the driver accurately knows the tire size.

For example, a vehicle type search method is a method of sequentially selecting an automobile manufacturer, a vehicle type, and a tire size. When choosing a size for a vehicle type, it is difficult to select the size because there are many different tire sizes for the same model. In addition, the size search method is a method of sequentially selecting the cross-sectional width, the flatness ratio, and the inch. If the exact size of the tire is not known, the product search can not be done. On the online tire sales site, it is possible to inform the driver of the size by image, text, video, etc. However, it is still difficult for the buyer to understand the size.

To search for products online, you must know your tire size correctly. If you do not know the tire size, you will not be able to buy the correct tire. However, most drivers do not know the size of the tire in their vehicle and feel it is difficult to buy online.

The present embodiments provide a tire information providing method using a portable terminal in an online area, for example, a PC, a mobile, and an application.

It is another object of the present invention to provide a tire information providing method which can provide information on currently mounted tire information and mountable goods in real time to assist in product selection and purchase.

The embodiments are not limited to the above-described problems, and various technical problems can be provided in the following detailed description.

According to another aspect of the present invention, there is provided a method of providing tire information using a mobile terminal, the method comprising: receiving a selection signal for an image search among search menus provided by a tire information providing system; Driving a camera module of the portable terminal in response to the received selection signal; Capturing a predetermined tire to obtain a tire image; Analyzing the obtained tire image; And providing the tire information according to the analysis result.

The tire information includes at least one of tire product information related to a tire same as or similar to the tire, damage information of the tire, and wear information of the tire.

The tire product information includes at least one of a manufacturer, a product name, a size, a product function, a manufacturing date, an offline price, an online price, a price comparison information, a factory price, a replacement determination information, review information, .

Wherein the analyzing step comprises the steps of: converting a circular tire portion in the tire image into a linear image; Extracting a recognition target character region in the date image; And recognizing the learning character most similar to the extracted recognition target character area as the character of the recognition target character area in the group of the learned learning characters.

The step of converting into the linear image includes extracting a circular tire portion in the image and rearranging linear images corresponding to a straight line perpendicular to the circumference of the circle or ellipse corresponding to the wheel region in the tire portion in the same direction Wherein the wheel portion is removed from the circular tire portion, the wheel portion is converted into the linear image, or the tire portion is removed from the circular tire portion, Extracting a certain region including the tire portion based on a guideline provided at the time of image photographing, and converting the extracted region into a linear image.

The step of transforming the image into a linear image may include: converting a circular tire part extracted based on a guideline provided at the time of photographing the tire image into a linear image; Calculating a boundary of the wheel region based on the intensity of the image of the first date format; And a step of re-extracting the circular tire portion in the tire image based on the boundary and performing secondary conversion to a linear image.

Wherein the step of calculating the boundaries of the wheel region includes: weighting a direction of each line of the image in the form of a rectangle; And obtaining a boundary line of each area based on intensities in the image of the date type; And calculating a shortest path connecting both ends of the path image by applying a weight according to the direction to the path image formed by the boundary lines of the respective regions.

Converting the shortest path to a circular image; Correcting the shortest path in the circular image to a circle or an ellipse; And re-extracting the in-image tire portion based on the corrected circle or ellipse; And converting the re-extracted tire portion into a straight tire image.

The step of extracting the character region may include extracting each character region using at least one of a texture, a pattern, and an edge in the image of the date format.

Wherein the extracting of the character region comprises: applying an entropy filter to the date image; Acquiring a binary image composed of regions of black and white using an adaptive threshold value for an image obtained by applying the entropy filter; Removing noise regions of a predetermined size or less from the binary image; And extracting the recognition target character region in the image from which the noise region is removed.

The step of recognizing the character includes: obtaining characteristic information by applying at least one filter to the character region to be recognized; And the learning character group is compared with the characteristic information of the character region to be recognized by applying the at least one filter to the recognition target character region to include the learning character having the most similar characteristic information in the recognition target character region And recognizing the character as a character.

Displaying at least one guide line on a screen of the portable terminal after the camera module is driven; And displaying the tire image obtained through the camera module on the screen,

Wherein the at least one guide line includes a circular first guide line and a circular second guide line having a smaller diameter than the first guide line in the first guide line,

And the shape of the guide line is adaptively modified according to the inclination of the portable terminal or the distance between the camera module and the tire.

Wherein the analyzing step comprises: searching for a comparison object image or comparison object data corresponding to the comparison object image in a database using at least one of the characters, numbers, and symbols located on the side of the tire image; And determining whether the tire image is damaged based on a difference between the tire image and the comparison object or a difference between the tire image data and the comparison object data.

Wherein the step of retrieving the comparison image comprises: converting a tire portion of a circle in the tire image into a linear image; Extracting a recognition target character region in the date image; Recognizing the learning character most similar to the extracted recognition target character region as the character of the recognition target character region in the pre-established learning character group; And searching the comparison object image in the database based on the recognized character.

Wherein the analyzing step comprises: detecting a line component in the tire image; Detecting a character region in the tire image; Removing line components located in the character region from the detected line components; And determining whether the damage is caused based on the line component remaining after the removing step.

Wherein the tire image is a moving image and generating an image of a three-dimensional shape of the tire based on the moving image; And measuring the wear of the tire tread based on the depth of the tread area in the image of the three-dimensional shape.

Wherein the generating the three-dimensional image comprises: separating the moving image into a plurality of still image units; Recognizing a pixel-by-pixel correspondence relationship between the plurality of still images; Determining a parameter including an angle and a distance at the time of photographing the moving picture through the pixel-by-pixel correspondence relationship; And generating depth information of the plurality of still images based on the parameters to generate a three-dimensional shape image for the tire.

Wherein the measuring the wear rate comprises: detecting a tread groove region and a surface region based on curvature analysis of the image of the three-dimensional shape; Determining a depth of the groove region from the detected surface region; And recognizing wear of the tire based on the detected depth.

Wherein the step of detecting the tread groove region and the surface region comprises: analyzing a curvature of each pixel of the three-dimensional shape image; Determining a region having a largest curvature among regions generated by connecting pixels having similar curvatures within a predetermined range and having mutual distances within a predetermined distance range; And detecting a tread groove region and a surface region separated on the basis of the region having the largest curvature.

According to another aspect of the present invention, there is provided a recording medium storing a program for causing a computer to execute a tire information providing method according to another embodiment.

The tire information providing method using the mobile terminal in the on-line area, for example, PC, mobile, and application according to the embodiment provides information on the currently installed tire information and the mountable product in real time, .

Also, the user can easily search for tires which are the same as or similar to the tires mounted on the vehicle, without knowing the tire size of the user, and purchase the tires cheaply by comparing the prices.

1 is a diagram illustrating a schematic structure of an overall system for providing tire information according to an embodiment.
FIG. 2 is a schematic block diagram of the portable terminal 110 and the tire information providing server 100 shown in FIG.
3 and 4 are illustrations for explaining a tire search method according to the related art.
5 is a flowchart for explaining a tire information providing method according to another embodiment.
6A is a schematic block diagram of the portable terminal 110 shown in FIG.
6B is a diagram illustrating an example of photographing a tire image through the portable terminal 110. As shown in FIG.
7 and 8 are views showing an example of a guideline that is modified according to photographing conditions according to another embodiment.
9 is a diagram showing a configuration of one example of the portable terminal 110 according to another embodiment.
10 is a flowchart illustrating an example of a method of acquiring a tire image through a portable terminal according to another embodiment of the present invention.
11 is a schematic block diagram of the tire recognition module 104 shown in FIG.
12 and 13 are views showing an example of a method of converting a circular tire portion according to yet another embodiment into a linear image.
14A to 14G are views showing an example of a method of correcting a linear image according to another embodiment.
FIGS. 14I through 14L are diagrams illustrating an example of a method of separating a date-shaped image according to another embodiment into regions.
FIG. 14M is a diagram illustrating an example of a method of correcting divided images according to another embodiment.
FIG. 15A is a diagram illustrating an example of a method of extracting a character region from an image separated by regions according to another embodiment. FIG.
FIG. 15B is a diagram illustrating an example of a method of extracting an image area in each character unit in a character area according to another embodiment.
FIG. 16 is a diagram illustrating an example of a method of comparing each character for tire character recognition according to another embodiment.
17 is a diagram showing a flow of an example of a tire recognizing method according to another embodiment.
FIG. 18 is a view showing a flow of an example of a conversion process to a linear image according to another embodiment.
FIG. 19 is a diagram showing a flow of an example of a correction method of a linear image according to another embodiment.
FIG. 20 is a diagram showing a flow of an example of a method of extracting a character region from a linear image according to another embodiment.
21 is a diagram showing a flow of an example of a tire character recognition method according to another embodiment.
22 is a schematic block diagram of the damage recognition module 105 shown in FIG.
23 is another block diagram of the damage recognition module 105 shown in FIG.
24 to 26 are flowcharts of a tire damage recognition method according to still another embodiment.
25 is a flowchart of a tire damage recognition method according to still another embodiment.
27 and 28 are views showing an example of a method of extracting an achromatic region from a tire image.
29 is a diagram showing an example of a method of extracting a black region in the achromatic region of FIG.
30 is a view showing an example of an image composed of black regions obtained by applying the method of FIG.
31 is a diagram showing an example of a method of extracting line components in the black region of FIG.
FIG. 32 is a diagram showing an example of a method of removing a line component of a character region in the line component image of FIG. 31;
FIG. 34 is a diagram showing an example of a method of recognizing a damage by comparing the line component of FIG. 31 with a comparison image of a database.
35 is a diagram showing a configuration of an embodiment of the wear rate measurement module 106. In Fig.
FIG. 36 is a flowchart showing an example of a method of measuring the tire wear rate according to another embodiment.
37 is a flowchart illustrating an example of a method of generating a three-dimensional image from a moving image for tire wear measurement according to another embodiment.
38 is a diagram showing an example of converting the two-dimensional coordinates of the plurality of still images into the spatial coordinates of the three-dimensional space.
39 is a view showing an example of an image of a three-dimensional shape acquired from a tire tread moving picture.
FIG. 40 is a flowchart showing an example of a concrete method of measuring tire tread wear from an image of the generated three-dimensional shape.
41 is a view showing an example of dividing a three-dimensional shape image by curvature magnitudes of pixels.
42 is a view showing an example in which a three-dimensional shape image is divided into a plurality of sections based on the direction and width of the tread groove region.
43 is a view showing an example of a three-dimensional shape image.
44 is a view showing an example of a method of grasping a tread groove depth from the three-dimensional shape image of FIG. 43;
45 is a view showing an example of correcting a three-dimensional shape image to an approximate plane.

Hereinafter, the tire information providing method according to the present embodiment will be described in detail with reference to the accompanying drawings.

1 is a diagram illustrating a schematic structure of an overall system for providing tire information according to an embodiment.

Referring to FIG. 1, the overall system includes a portable terminal 110, a communication network 120, and a tire information providing server 100.

The portable terminal 110 is a user terminal including a smart phone, a PC, a notebook, a wearable computer, and the like, and is hereinafter referred to as a portable terminal 110.

The tire information providing server 100 provides tire information to the portable terminal 110 connected through the communication network 120. [ Here, the tire information may include the same tire product information as the tire currently mounted on the user's vehicle, damage information of the currently mounted tire, wear information of the currently mounted tire, and the like. The tire product information may include a manufacturer, a product name, a size, a product function, a manufacturing date, an offline price, an online price, and the like.

The tire information providing server 100 provides tire information to the portable terminal 110, provides tire purchasing and payment on-line, ships the purchased tire to a mounting shop desired by the user, It provides a service to replace tires. The concrete configuration of the tire information providing server 100 will be described later with reference to FIG.

FIG. 2 is a schematic block diagram of the portable terminal 110 and the tire information providing server 100 shown in FIG.

The mobile terminal 100 connected to the tire information providing server 100 may be provided with a vehicle type search 111, a size search 112 and an image search 113 as a search menu for tire search.

As shown in FIGS. 3A to 3D, the vehicle type search 111 is a method in which a user sequentially selects an automobile manufacturer, a vehicle type, and a tire size. However, when selecting a size for such a vehicle search method, it is difficult to select the size because there are many different tire sizes for the same vehicle model.

The size search 112 is a method of sequentially selecting a cross-sectional width, a flatness ratio, and an inch as shown in Figs. 4A to 4D. The size search method can not be selected unless the size of the tire is known precisely, which is more difficult than the method of searching for a car.

The vehicle type search and size search methods shown in FIGS. 3 and 4 are methods in which a product can be searched by knowing the tire size as described above.

When the user photographs a tire mounted on his / her vehicle with his / her portable terminal 110, the image search 113 transmits a live view image of the camera module to the tire information providing server 100, (100) analyzes the transmitted tire image, recognizes the tire, searches for the same or similar tire, and provides the tire product information to the user as tire information. It is also possible to provide not only the tire product information but also the damage information of the currently mounted tire or the wear degree information of the tire. Here, the tire product information includes at least one of manufacturer, product name, size, product function, manufacturing date, offline price, online price, price comparison information, factory price, replacement determination information, review information, preference information, But is not limited thereto.

With reference to the image search 113, this will be described with reference to FIG.

5 is a flowchart for explaining a tire information search method according to another embodiment.

Referring to FIG. 5, in step 500, the user selects an image search. Here, the user selects an image search through one of the search interfaces provided by the tire information providing server 100 through the portable terminal 110.

In step 502, the camera module included in the portable terminal 110 is driven. The camera module is automatically initialized according to the selection of the user's image search 113, and the camera is ready for shooting.

In step 504, when the camera module is driven, the user uses his or her portable terminal 110 to take a tire mounted on the vehicle as a subject, and obtains a live view image of the tire. However, the present invention is not limited to this, and may be a captured image obtained by photographing a tire by clicking a photographing button or a photographing icon by a user.

In step 506, the captured tire image (live view image or captured image) is transmitted to the tire information providing server 100. However, in the case where an application for providing tire information is mounted on the portable terminal 110, a photographed tire image is provided to the application to provide image analysis and tire information Of course.

At step 508, the tire image is analyzed and recognized. The tire image analysis includes recognizing the characters in the tire to recognize the tire, recognizing the damage of the tire, or measuring the wear of the tire.

In step 510, tire information is provided. As shown in Figs. 3d and 4d, it is possible to provide a list of goods for the same tire as the tire mounted on the user's vehicle, or to provide tire damage information, tire wear information.

2, the tire information providing server 100 includes a control module 101, a vehicle type search module 102-1, a size search module 102-2, an image search module 102-2, 3), a connection interface 103 for controlling network connection with user terminals, a tire recognition module 104 for recognizing tires, a tire damage recognition A module 105, and a tire wear measurement module 106 for measuring the wear of the tire. Here, the search interface 102 and the analysis modules 104 to 106 may be implemented as an application, and may be implemented in the portable terminal 110 and implemented.

The tire information providing server 100 may include a user DB 107-1 and a tire DB 107-2 and may include a payment module 108 and a delivery module 109 responsible for tire settlement of a user . Here, although the tire information providing server 100 includes the payment module 108 and the delivery module 109, it is needless to say that the payment / delivery service can be performed to the user in cooperation with the external payment / delivery system .

6A is a schematic block diagram of the portable terminal 110 shown in FIG.

The portable terminal 110 includes a sensing unit 140, an A / V input unit 160, and a memory 170 in addition to the user input unit 131, the output unit 132, the control unit 130, and the communication unit 150 .

The user input unit 131 may mean a means for the user to input a selection for tire search. For example, the user input unit 131 may be a key pad, a dome switch, a touch pad (a contact type capacitance type, a pressure type resistive type, an infrared detection type, a surface ultrasonic wave conduction type, A tension measuring method, a piezo effect method, etc.), a jog wheel, a jog switch, and the like, but is not limited thereto.

The output unit 1320 may output an audio signal or a video signal or a vibration signal and the output unit 132 may include a display unit 133, an acoustic output unit 134, and a vibration motor 135 have.

The display unit 133 may be a liquid crystal display, a thin film transistor-liquid crystal display, an organic light-emitting diode, a flexible display, a three-dimensional display A 3D display, and an electrophoretic display.

The sound output unit 134 can output audio data received from the communication unit 150 or stored in the memory 170. [ The sound output unit 134 may include a speaker, a buzzer, and the like.

The vibration motor 135 can output a vibration signal. For example, the vibration motor 135 may output a vibration signal corresponding to an output of audio data or video data (e.g., a call signal reception tone, a message reception tone, etc.). In addition, the vibration motor 135 may output a vibration signal when a touch is input to the touch screen.

The control unit 130 can generally control the overall operation of the portable terminal 110. For example, the control unit 130 may include a user input unit 131, an output unit 132, a sensing unit 140, a communication unit 150, an A / V input unit 160 ) Can be generally controlled. Also, the controller 130 may control the components in the mobile terminal 110 to perform the tire search operation of the mobile terminal 110.

Specifically, when the user selects the image search, the control unit 130 drives the camera module 161 to photograph the tires. The captured tire image may be transmitted to the tire information providing server 100 or may be provided to the tire analysis module in the control unit 130 to analyze the tire image when the image is implemented as an application.

The sensing unit 140 may sense the state of the portable terminal 110 or the surrounding state, and may transmit the sensed information to the controller 130. [

The sensing unit 140 includes a magnetism sensor 141, an acceleration sensor 142, a temperature / humidity sensor 143, an infrared sensor 144, a gyroscope sensor 145, (GPS) 146, an air pressure sensor 147, a proximity sensor 148, and an RGB sensor (illuminance sensor) 149. However, the present invention is not limited thereto. The function of each sensor can be intuitively deduced from the name by those skilled in the art, so a detailed description will be omitted.

The communication unit 150 may include one or more components for communicating between the portable terminal 110 and the tire information providing server 100. For example, the communication unit 150 may include a short-range communication unit 151, a mobile communication unit 152, and a broadcast reception unit 153.

The short-range wireless communication unit 151 includes a Bluetooth communication unit, a BLE (Bluetooth Low Energy) communication unit, a Near Field Communication unit, a WLAN communication unit, a Zigbee communication unit, IrDA, an infrared data association) communication unit, a WFD (Wi-Fi Direct) communication unit, an UWB (ultra wideband) communication unit, an Ant + communication unit, and the like.

The mobile communication unit 152 can transmit and receive a radio signal to at least one of a base station, an external terminal, and a server on a mobile communication network. Here, the wireless signal may include various types of data depending on a voice call signal, a video call signal, or a text / multimedia message transmission / reception.

The broadcast receiving unit 153 can receive broadcast signals and / or broadcast-related information from outside through a broadcast channel. The broadcast channel may include a satellite channel and a terrestrial channel. The portable terminal 110 may not include the broadcast receiving unit 153 according to the embodiment.

The communication unit 150 may transmit the tire image to the tire information providing server 100 and receive the tire information from the tire information providing server 100. [

The A / V (Audio / Video) input unit 160 is for inputting an audio signal or a video signal, and may include a camera 161 and a microphone 162. The camera 161 can obtain an image frame such as a still image or a moving image through the image sensor in a video communication mode or a photographing mode. The image captured through the image sensor can be processed through the control unit 130 or a separate image processing unit (not shown).

The image frame processed by the camera 161 can be stored in the memory 170 or transmitted to the outside through the communication unit 150. [ The camera 161 may be equipped with two or more cameras according to the configuration of the terminal.

The microphone 162 can receive an external sound signal and process it as electrical sound data. For example, the microphone 162 may receive acoustic signals from an external device or speaker. The microphone 162 may use various noise reduction algorithms to remove noise generated in receiving an external sound signal.

The memory 170 may store a program for processing and controlling the control unit 130 and may store data input to or output from the portable terminal 110. [

The memory 170 may be a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory), a RAM (Random Access Memory) SRAM (Static Random Access Memory), ROM (Read Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory) , An optical disc, and the like.

The programs stored in the memory 170 may be classified into a plurality of modules according to their functions, for example, a UI module 171, a touch screen module 172, a notification module 173, .

The UI module 171 can provide a specialized UI, a GUI, and the like, which are interlocked with the portable terminal 110 for each application. The touch screen module 172 senses a touch gesture on the user's touch screen and can transmit information on the touch gesture to the control unit 130. [ The touch screen module 172 according to some embodiments may recognize and analyze the touch code. The touch screen module 172 may be configured as separate hardware including a controller.

Various sensors may be provided in or near the touch screen to sense the touch or near touch of the touch screen. An example of a sensor for sensing the touch of the touch screen is a tactile sensor. A tactile sensor is a sensor that detects the contact of a specific object with a degree or more that a person feels. The tactile sensor can detect various information such as the roughness of the contact surface, the rigidity of the contact object, and the temperature of the contact point.

In addition, a proximity sensor is an example of a sensor for sensing the touch of the touch screen.

The proximity sensor refers to a sensor that detects the presence or absence of an object approaching a predetermined detection surface or a nearby object without mechanical contact using the force of an electromagnetic field or infrared rays. Examples of proximity sensors include a transmission type photoelectric sensor, a direct reflection type photoelectric sensor, a mirror reflection type photoelectric sensor, a high frequency oscillation type proximity sensor, a capacitive proximity sensor, a magnetic proximity sensor, and an infrared proximity sensor. The user's touch gestures can include tap, touch & hold, double tap, drag, panning, flick, drag and drop, swipe, and the like.

The notification module 173 may generate a signal for notifying the occurrence of an event of the portable terminal 110. Examples of events generated in the portable terminal 110 include call signal reception, message reception, key signal input, schedule notification, and the like. The notification module 1731 may output a notification signal in the form of a video signal through the display unit 133 or may output a notification signal in the form of an audio signal through the sound output unit 134, It is possible to output a notification signal in the form of a vibration signal.

6B is a diagram illustrating an example of photographing a tire image through the portable terminal 110. As shown in FIG.

Referring to FIG. 6B, on the screen of the portable terminal 110, an image to be photographed by the camera is displayed, and at least one or more guide lines 600 and 610 for assisting the photographing of the tire image are displayed.

For example, the guidelines 600 and 610 may include a first guide line 600 corresponding to the entire tire size and a second guide line 610 corresponding to the wheel portion of the tire. The first and second guide lines 600 and 610 may have two circular or elliptical shapes with different diameters and the second guide line 610 may be positioned within the first guide line 600.

As another example, the guidelines 600 and 610 may be variously modified such as a polygonal shape such as a triangle, a rectangle, a pentagon, or a shape in which only a part of a circle or a polygon appears. Further, the number of guide lines 600 and 610 may be one or three or more.

The user can take a tire image in accordance with the guidelines 600 and 610 displayed on the screen of the portable terminal 110. At this time, the shapes of the guide lines 600 and 610 may be changed according to the distance between the portable terminal 610 and the subject or the tire image photographing angle, which will be described with reference to FIGS.

Referring to FIG. 7, the curvature of the guideline displayed on the screen of the terminal 110 is changed according to the photographing angle of the camera, that is, the tilting degree of the portable terminal 110. For example, when the user takes a tire image in front of the tire according to the height of the tire, the slope of the terminal 110 is substantially zero, and the shape of the guide lines 700 and 710 is a circle in this case.

On the other hand, when the terminal 110 is tilted forward, such as when a camera takes a tire image in a downward direction, the image of the tire taken by the camera becomes an elliptical shape, so that the guide lines 720 and 730 also fit It is changed into an elliptic shape.

Referring to FIG. 8, when the distance between the portable terminal 110 and the tire, that is, the distance from the subject, is small, the sizes of the guide lines 800 and 810 are increased, As the distance increases, the size of the guide lines 820 and 830 becomes smaller.

As described above, the curvature, size, and position of the guide line are adaptively modified according to the photographing angle of the terminal and the distance to the subject, and displayed on the terminal screen. In other words, the shape and the like of the tire image photographed according to the photographing conditions such as the direction of taking a tire image in the terminal, the distance, the illumination, and the like are different. Therefore, the mobile terminal 110 controls the shape of the guide line according to the shape do.

Although the present embodiment describes the deformation of the guiding line according to the distance between the angle of the terminal and the subject, the guiding line deformation method according to various other imaging conditions can be applied.

9 is a diagram showing a configuration example of a processor of the portable terminal 110. As shown in Fig.

9, the portable terminal 110 includes a guide line display unit 900, an image acquisition unit 910, and a transmission unit 920. [

The guide line display unit 900 displays a guide line for photographing a tire image on the screen. The guideline may be a circular or elliptical shape in which the size or the curvature is deformed according to the inclination of the terminal or the distance to the subject tire, but is not limited thereto. The image acquiring unit 910 acquires the tire image through the camera. For example, as shown in FIG. 6B, the user aligns the guide lines 600 and 610 indicated by two ellipses on the terminal screen with the wheel portion of the target tire and the outside of the tire. The transmitting unit 920 transmits the image including the tire portion taken by the image obtaining unit 910 directly or through the third device to the tire information providing server 100. The transmitter 920 may be omitted if the tire information providing server 100 is constructed as a kind of application and implemented in the terminal.

If the guideline is fixed in the terminal screen instead of adaptively changing its shape according to the terminal status and the tire information providing server 100 knows information such as the position and size of the fixed guideline in the terminal in advance, The transmitting unit 920 may not transmit information on the size, curvature, position, etc. of the guideline to the tire information providing server 100. [ However, if the guide line is changed according to the terminal status, or if the tire information providing server 100 does not have the advance information on the guide line, the transmitter 920 transmits information such as position, size, To the tire information providing server 100 together with the image.

10 is a flowchart illustrating an example of a method of acquiring a tire image through a portable terminal according to another embodiment of the present invention.

Referring to FIG. 10, in step 1000, the terminal recognizes an angle of view, a distance to a subject, and the like. The photographing angle can be obtained by grasping the inclination of the terminal through various sensors capable of detecting a tilt such as a gyro sensor built in the terminal. The distance to the subject can be determined through an auto focus algorithm or various distance measuring sensors.

In step 1002, the portable terminal 110 controls the size, the curvature, and the position of the guide line displayed on the screen according to the photographing angle, the distance to the subject, and the like. According to an embodiment, the guideline may be fixed and not changed according to the state of the terminal.

In step 1004, the terminal captures an image including the tire portion via the camera. In shooting, the user photographs the guide line displayed on the screen in correspondence with the tire portion, so that the tire information providing server 100 can more easily identify and analyze the tire portion.

In step 1006, the terminal transmits the photographed image to the tire information providing server 100 directly or via the third device. If the shape, size, or curvature of the guide line is adaptively changed according to the terminal state, the terminal can transmit information such as the size and position of the guide line to the tire information providing server 100 together with the tire image .

11 is a schematic block diagram of the tire recognition module 104 shown in FIG.

11, the tire recognition module 104 includes an image conversion unit 1110, a correction unit 1120, an area division unit 1130, a character region extraction unit 1140, a character recognition unit 1150, (1160). The tire recognition module 104 according to the embodiment allows tire side characters that are not clearly distinguished in black and white, unlike a general paper document. Therefore, the user can easily acquire information about the tire by photographing the tire lateral image using a smart phone or the like. In addition, it is possible to provide a guideline for shooting a tire image, thereby facilitating the tire recognition process.

The image converting unit 1110 converts the tire portion of the circle in the image into a date-shaped image. More specifically, the image transformation unit 1110 estimates the tire wheel part boundary using a point having a clear boundary between the wheel part in the image and the remaining tire part, and estimates the tire wheel part boundary using the estimated wheel part boundary, And convert the extracted tire portion into an image in the form of a date. The image converting unit 1110 may perform various image processing processes including histogram equalization and the like for a linear image in order to obtain a clearer image.

The present embodiment illustrates the case of converting a tire portion from which a wheel is removed from a tire into a straight image, but the present invention is not limited thereto. The present invention is not limited to this, It is possible to convert a tire portion including a part of the wheel portion into an image of a straight shape.

In the following description, however, the case of converting the tire portion removed from the tire wheel portion into a straight image is mainly described.

For example, the image converting unit 1110 may extract the tire portion 630 located between the two circular guide lines 600 and 610 as shown in FIG. 12, and convert the tire portion 630 into a straight image. The image conversion in the form of a date can be applied to the method shown in Fig. However, the present embodiments are not limited to the method shown in FIG. 13, and circular images can be laid out in a date form by using various conventional image processing methods.

In the case of extracting the tire portion using the guidelines 600 and 610, not only a certain portion of the wheel portion is extracted together with the upper portion of the wheel portion as shown in the upper diagram 1460 of FIG. 14G but the tire portion is not straightened properly, And correcting the tire portion to obtain the lower figure 1461 of the piece 14g. This correction process is performed by the correction unit 1120, and the correction unit 1120 may be omitted according to the embodiment.

The correcting unit 1120 corrects the wheel part boundaries used for obtaining the image in the form of a date in the image converting unit 1110, and re-extracts the tire part in the image based on the corrected boundary to generate a straight image.

For example, the corrector 1120 grasps the tire wheel part boundaries based on intensities of the date-shaped images generated by the image transform unit 1110. The boundary between the wheel area and the remaining tire area in the straight image is generally on a straight line in the horizontal direction. Accordingly, the correcting unit 1120 obtains boundary lines for a plurality of regions classified on the basis of intensities in a linear image, calculates a shortest path connecting both ends of a path-shaped path image composed of boundary lines, By taking into consideration the weight according to the weight. The calculated shortest path corresponds to the tire wheel part boundary. Here, the shortest path is not the shortest path but the shortest path considering the weight. For example, if the length of the first path is 10, the weight is 0.8, the length of the second path is 11, and the weight is 0.5, the weighted length of the first path is 10 * 0.8 = 8, The length is 11 * 0.5 = 5.5, and the second path is selected as the shortest path. A detailed description thereof will be given in Figs. 14A to 14G.

The area dividing unit 1130 divides the image in the form of a line into at least one area using a material, a pattern, and an edge. For example, the tire side is divided into tread regions, regions where large characters exist, regions where small characters exist, materials and patterns are different from each other, so that boundaries of these different materials are extracted to separate the regions. Various methods of distinguishing each region in an image may exist, and an example thereof will be described with reference to Figs. 14H to 14M. Although the present embodiment shows a configuration of an area dividing section 1130 that divides a number of areas including a character string into a linear image so as to facilitate extraction of a character area, May be omitted.

The character region extracting unit 1140 extracts a recognition target character region in which characters exist in the respective regions divided by the region dividing unit 1130. [ As another example, when the region dividing unit 1130 is omitted, the character region extracting unit 1140 extracts a recognition target character region in which characters exist in the entire date image. Here, the letters indicate alphabets, Hangul characters, numbers, symbols, images, and the like that display tire information.

The character region extracting unit 1140 extracts a character using a variety of image analysis methods such as intensities analysis, histogram analysis, edge analysis, and the like, And extracts the recognition target character area. Here, the recognition target character area is not a character itself but a certain area including characters. For example, the character region extracting unit 1140 may first extract an area composed of a plurality of characters as shown in Fig. 15, and extract each character region in the first extracted region as a character region to be recognized.

The character recognition unit 1150 determines what characters exist in each recognition target character area extracted by the character area extraction unit 1140. [ Since the characters in the tire image and the surrounding background color are not distinguished from each other in black and white, it is difficult to recognize a character by a general image recognition method. Therefore, the character recognition unit 750 constructs a learning database 1160 including characteristic information about the learning character group in advance, and then outputs the character most similar to each character extracted by the character area extraction unit 1140 to the learning database 1160 , The character of the recognition target character area is grasped.

The user may include a learning database 1160 for storing at least one or more pieces of feature information obtained by separating a learning region including each character in a tire photographing image and applying at least one filter to each learning region together with character information of the learning region Build. For example, the user extracts an image area of a predetermined size including the learning character "M" in the tire side image, and then stores one or more pieces of feature information obtained by applying the filter to the area in the learning database 1160. Therefore, if the feature information of the recognition target character area extracted by the character area extraction unit 1140 is most similar to the feature information of "M" preliminarily stored in the learning database 1160, the character recognition unit 1150 extracts And recognizes that the character existing in the area is 'M'. An example of a method of recognizing characters by comparing feature information using a filter is shown in Fig.

12 and 13 are views showing an example of a method of converting a circular tire portion according to yet another embodiment into a linear image.

12 and 13, the tire recognition module 104 converts a circular image 630 located between the first and second guide lines 600 and 610 provided at the time of image shooting into a tire image 1200 in the form of a straight line . At this time, the tire recognition module 104 generates linear images 1320 and 1320 corresponding to a predetermined number (for example, N) of straight lines arranged in a direction perpendicular to the circumference of the second guide line 610 corresponding to the wheel region -m) are rearranged in one direction to generate a straight-line tire image 1200.

For example, N straight lines that are perpendicular to the circumference are set from 0 degrees to 360 degrees of the circular second guide line 610, and a straight line image consisting of pixels in the tire extracting image 630 corresponding to N straight lines (1320 and 1320-m) are sequentially arranged in the same direction. The first straight line image 1320 of 0 degree and the mth straight line image 1320-m of the A degree are arranged in the same direction on the same line as in Fig.

14A to 14G are views showing an example of a method of correcting a linear image according to another embodiment.

Referring to FIG. 14A, the tire recognition module 104 applies a direction filter capable of giving a weight according to a direction to a linear image. 12, since the boundary of the wheel region is close to the transverse direction, that is, the 0 degree direction, the tire recognition module 104 has a directional filter of 0 degree direction to give a higher weight to the transverse direction, Is applied to a date image, and when the size of the weight is expressed by color, the image 1400 is displayed as shown in FIG. 14A.

For example, the tire recognition module 104 may assign a weight 1 to the 0 degrees (or 360 degrees) direction for the lines appearing in the linear image, a weight 0 to the 90 degrees (or 180 degrees) direction, The weights of 0 and 1 are given according to the angle.

Referring to FIG. 14B, the boundaries of the respective regions, which are classified based on intensities in the image of the tire recognition module 104, are obtained. For example, if the intensities of each pixel of the image in the form of a date are expressed (1430), ridges (1431, 1432, 1433) appear between the regions as shown in FIG. The tire recognition module 104 calculates a path image 1410 of a complex labyrinth shape represented by the boundary lines made up of such ridges as shown in FIG. 14B. The tire recognition module 104 may use a wartershed algorithm as an example of an algorithm for grasping the ridgeline portion classified based on intensities.

14C, the tire recognition module 104 obtains the image 1420 of FIG. 14C, which reflects the weight of the image in the form of a line image as shown in FIG. 14A in the path image 1410 of FIG. 14B, The shortest path that reflects the weight. For example, the tire recognition module 104 may combine the images of Figures 14A and 14B to obtain the image of Figure 14C.

14E, since the left and right ends correspond to each other, the tire recognition module 104 uses the point where the left start point and the right end start point are the same, And the paths are identified by applying various conventional path search algorithms. At this time, the tire recognizing module 104 sets a route that can be easily reached as the weight is higher, and calculates the route 1440 that can be passed most easily.

Referring to Fig. 14F, the tire recognition module 104 applies the path obtained in Fig. 14E to the image in the shape of a rectangle in Fig. However, a step 1450 similar to a step may appear in the path due to the smoothing effect of the direction filter in FIG. 14A or the like.

In this case, the tire recognizing module 104 may further perform a process of correcting the path 1510 to a smooth line by removing a phenomenon such as a step. For example, the tire information recognition module 104 may apply a dynamic contour model that considers the elasticity between each pixel and the gradient force of the external image to remove the staircase of the path 1510 have. In addition, since the boundary of the wheel portion appearing in the actual tire image is circular or elliptical, the tire recognition module 104 converts the diagonal image 1450 including the path 1510 into a circular image, Or an ellipse.

Referring to FIG. 14G, when a step 1510 representing a tire wheel boundary as shown in FIG. 14C or a stepped phenomenon as shown in FIG. 14F appears, the corrected path is applied again to the received image to extract a tire portion excluding the wheel portion, And converts the extracted portion into an image 1461 in the form of a date. It can be seen that the linear image 1461 after correcting the wheel region boundary more accurately distinguishes the tire portion than the straight image 1460 before correcting the wheel region boundary.

FIGS. 14H to 14L are views showing an example of a method of separating the image in the form of a line, according to another embodiment.

14H, the tire recognition module 104 uses the fact that the complexity is large around the character area in the image and the pattern is small and the complexity is small in the area where the pattern or the feature is not, Apply a filter to increase the difference between the text area and the rest.

14I, the tire recognition module 104 converts the linear image 1480 into a binary image 1481 of black and white. In one example, the tire recognition module 104 may convert the linear image 1480 into a binary image 1481 using an adaptive threshold. In this case, the tire recognition module 104 determines whether or not the intensities of a predetermined number of pixels (for example, 3X3 pixels) in the kernel are adaptively changed according to standard deviations of pixel values in the kernel, The pixel is determined to be white and the other side is determined to be black so that a binary image 1810 composed of black and white can be obtained.

Referring to FIG. 14J, the tire recognition module 104 removes the noise region of the binary image 1481 of FIG. 14I to produce an image 1490 as shown in FIG. 14J. For example, the tire recognition module 104 may obtain an image 1490 of FIG. 14J by removing a white area smaller than a predetermined size in the binary image 1481 of FIG. 14I.

Referring to FIG. 14K, the tire recognition module 104 applies a shortest path algorithm to the noise-removed binary image 1500 to calculate at least one path 2010,2020 across the two ends.

Referring to FIG. 14L, the tire recognition module 104 applies the paths 2010 and 2020 obtained in FIG. 14K to the linear image 1600 to determine a tread area and a region in which large characters exist, a region in which small characters exist .

FIG. 14M is a diagram illustrating an example of a method of correcting divided images according to another embodiment.

Referring to FIG. 14M, the tire recognition module 104 performs a background correction process such as eliminating the effect of the external light source in order to more clearly distinguish the character regions included in the images of the divided regions 2200. For example, the tire recognition module 104 calculates images 2210 and 2220 sequentially applied to each of the divided regions 2200 by applying an image processing process such as a motion blur and an entropy filter.

FIG. 15A is a diagram illustrating an example of a method of extracting a character region from an image divided into regions according to another embodiment, FIG. 15B is a diagram illustrating an example of extracting an image region in each character unit in a character region according to another embodiment And the like.

Referring to FIG. 15A, since the tire recognition module 104 has many edges around characters, each character region can be extracted based on feature points acquired on the basis of edges. Referring to FIG. 15B, the tire recognition module 104 extracts at least one or more words included in each character region in units of characters.

FIG. 16 is a diagram illustrating an example of a method of comparing each character for tire character recognition according to another embodiment.

16, at the time of building a system, at least one or more feature information 1810, 1820 obtained by separating a character region located in a tire side image into a learning object region 1800 and applying a filter to the learning object region 1800, To the learning database. Here, a garber filter or a Haar-like filter can be used as the filter.

For example, the tire recognition module 104 analyzes the tire image received from the terminal, extracts a recognition target character area 1900 including 'A', and adds at least one or more A filter is applied to obtain one or more feature information (1910, 1920). The tire recognition module 104 compares the feature information 1910 and 1920 of the recognition target character area 1900 with the feature information previously constructed for each learning character in the learning database and obtains the most similar feature information 1810 and 1820 And recognizes the learned learning character 'A'.

17 is a diagram showing a flow of an example of a tire recognizing method according to another embodiment.

Referring to FIG. 17, in step 2000, a tire image is received. In step 2010, the tire recognition module 104 converts the received tire image portion into an image in the form of a date. According to the embodiment, it is possible to convert only the tire portion from which the tire wheel portion is removed into an image in the form of a straight line, or convert an entire tire portion including the tire wheel portion into a straight line image. The tire recognizing module 104 may use the guideline provided at the time of image photographing to remove the wheel part or the wheel part boundary obtained by applying various image processing methods.

In step 2020, the tire recognition module 104 extracts each character region by using a pattern, texture, edge, or the like in the image in the form of a line. The tire recognizing module 104 may divide the image in a line shape into a plurality of regions, and then extract a recognition target character region including characters existing in each region. At this time, various image processing processes such as entropy filter, motion blur, and the like can be performed in order to extract a clear character region.

In step 2030, the tire recognition module 104 compares the recognition target character area with a learning database to determine what character is to be recognized. The learning database stores one or more feature information obtained by applying various filters to each character region existing in the tire side image. Accordingly, the learning database having the feature information most similar to the one or more feature information obtained by applying the filter to the extracted recognition target character region is searched in the learning database, thereby recognizing the character of the extracted character.

The tire recognition module 104 recognizes each character existing in the tire image, and recognizes the tire information such as the type and size of the tire by referring to the database in which the tire information is stored, and provides the obtained tire information to the terminal.

FIG. 18 is a view showing a flow of an example of a conversion process to a linear image according to another embodiment.

Referring to FIG. 18, in step 2100, the tire recognition module 104 removes the tire wheel portion in the image. The tire recognition module 104 may use the guideline provided at the time of image photographing to remove the wheel portion.

In step 2110, the tire recognition module 104 converts the circular tire portion from which the wheel portion has been removed into an image in the form of a line. There can be various ways to convert a circular image to a date format.

When a guideline is used to convert an image in the form of a date, a certain portion of the wheel portion may be included in the date image.

In step 2120, a wheel part boundary is extracted from the linear shape image to remove a more accurate wheel part, a tire part is extracted from the image again using the wheel part boundary, and a correction process is performed to convert the tire part image into a linear shape image. The detailed configuration of the correction process is described in Fig.

FIG. 19 is a diagram showing a flow of an example of a correction method of a linear image according to another embodiment.

Referring to Fig. 19, in step 2200, the tire recognition module 104 acquires a first image to which a directional filter is applied to a linear image. 12A and 12B, since the boundary of the wheel part is close to the horizontal direction, a direction filter that gives a higher weight value toward the 0 degree direction is applied to the linear image 1200, 1 image 1400 is obtained.

In step 2210, the tire recognition module 104 obtains a second image, which is a boundary line of each area divided on the basis of intensities in a linear image. As shown in FIG. 14D, the tire recognition module 104 uses the water-shed algorithm or the like to display the ridgelines as boundary lines, as shown in FIG. 14B And acquires a second image 1410 such as < / RTI >

In step 2220, the tire recognition module 104 calculates the shortest path considering the weight by applying the path search algorithm to the maze-like path made up of the weighted boundaries by combining the first image with the second image. The shortest path thus obtained corresponds to the boundary of the tire wheel portion.

In step 2230, the tire recognition module 104 converts the shortest path existing in the linear image into a circular image before conversion into a date form, and corrects the shortest path existing in the circular image to a circle or an ellipse.

In step 2240, the tire recognition module 104 re-extracts the tire portion in the original image based on the circle or ellipse created by correcting the shortest path, and converts it into a linear image.

FIG. 20 is a diagram showing a flow of an example of a method of extracting a character region from a linear image according to another embodiment.

Referring to FIG. 20, in steps 2300 and 2310, the tire recognition module 104 performs a pre-processing step such as an entropy filter, a motion blur, and the like on a linear image to obtain a clear image. Depending on the embodiment, the tire recognition module 104 may perform several other preprocessing steps or omit the preprocessing step.

In step 2320, the tire recognition module 104 converts the date-type image in which the preprocessing step has been completed into a binary image composed of black and white. The tire recognition module 104 may obtain a binary image by applying an adaptive threshold.

In step 2330, the tire recognition module 104 removes the noise region of the binary image, i.e., the region smaller than the predetermined size. An example of the noise canceled state is shown in FIG. 14J.

In step 2340, the tire recognition module 104 applies a path search algorithm to the noise-removed binary image to calculate at least one path connecting both ends, and then in step 2350, the path 2010202020 ) Is applied to the date image to divide it into at least one area.

In step 2360, the tire recognition module 104 extracts a recognition target character area including characters in each divided area. Unlike a typical paper document, the character portion in a tire image can not be recognized by applying a general character recognition algorithm since the distinction between the background and black and white is not clear. Therefore, the tire recognizing module 104 extracts a region having many edges as a portion where a character exists, based on the characteristic that many edges exist in the case of characters. As shown in FIG. 15, for example, between the letters 'M' and 'I' in a certain area having a large number of edges as shown in FIG. 15 is a flat area without edges. Therefore, It is possible to extract the recognition target area divided for each character.

21 is a diagram showing a flow of an example of a tire character recognition method according to another embodiment.

Referring to FIG. 21, at step 2400, a learning database is first constructed. The learning database stores at least one feature information obtained by applying at least one filter to a learning region containing each character extracted from the tire image.

In step 2410, the tire recognition module 104 analyzes one or more images received from the terminal and applies one or more predetermined filters to the extracted recognition target area to calculate one or more pieces of feature information. In step 2420, the tire recognition module 104 compares the feature information of the recognition target character area with the feature information stored in the learning database for each learning character, and grasps the learning character having the most similar characteristic information as the character of the recognition target character area. For example, if the feature information of the recognition target character region is similar to the feature information of the region including the learning character 'A' in the learning database, the tire recognition module 104 determines that the character existing in the recognition target character region is 'A' .

However, some characters may not be recognized correctly due to tire wear or contamination. In this case, in steps 2430 and 2440, the tire recognizing module 104 compares the recognition result of the character string of the specific area with the character string stored in the learning database in advance, and obtains the most similar character string from the character Heat. For example, the tire recognition module 104 erroneously recognizes some characters such as 'MIO ELIN' and the like, due to 'MICHELIN' which is originally present in the tire image, tire wear or contamination, May not be recognized at all. In this case, the tire recognizing module 104 recognizes the character string 'MICHELIN' most similar to "MIO ELIN" among the strings stored in the learning database as a character string of a tire to be recognized.

22 is a schematic block diagram of the damage recognition module 105 shown in FIG.

22, the damage recognition module 105 includes a DB search unit 2210, a comparison object image database 2220, a comparison unit 2230, and a damage recognition unit 2240. The tires suffer damage (3400) such as cracks as shown in Fig. 34 due to various causes during the manufacturing process or use. Since there are letters and various patterns and patterns on the tire side, it is not easy to distinguish cracks and other damages through a general image analysis process. Accordingly, the damage recognition module 105 according to the embodiment allows a user to recognize a tire damage, for example, a crack, through an image photographed through a camera without directly checking with the eyes. Accordingly, the user can easily recognize whether or not the tire is damaged by photographing the tire image using a camera or a smart phone with a built-in camera function. In addition, when tire damage is present and tire replacement is required, the user can be provided with information about the tire.

The DB search unit 2210 searches the database for a comparison target image using at least one of letters, numbers, and symbols located on the side of the tire image.

For example, the DB search unit 2210 recognizes the characters existing on the side of the tire image and recognizes the tire information, and then searches the database based on the detected tire information to extract the comparison target image. At this time, the DB search unit 2210 can use the tire information recognized by the tire recognition module 104. [

For example, the DB search unit 2210 receives the tire information directly from the terminal, grasps the characters printed on the side of the tire through an image processing method other than the method described by the tire recognition module 104, Or to allow a user to select a specific tire from a server connected through a terminal, or the like.

As another example, the DB search unit 2210 can extract a comparison object image having a characteristic similar to that of the tire image from the database 2220 without grasping the tire information. After storing at least one or more pieces of feature information extracted by applying a filter capable of extracting feature information, such as a Gabor filter and a low-level filter, to the tire image in advance in the database together with the comparison object image, the DB search unit 2210 The database 2220 can extract the comparison object image having the feature information most similar to the feature information obtained by applying the filter to the received tire image. In this case, for more accurate comparison with the feature information of the comparison object, the DB search unit 2210 can perform the process of correcting the size and direction of the received tire image to predetermined sizes and directions before the feature information extraction .

The database stores the comparison target image for each tire in advance. Here, the image to be compared is an image in which no foreign substance is observed, such as a tire image at the time of delivery of a tire. The database maps the tire information and stores the comparison target image. As another example, in the case of searching a comparative image based on the feature information, the database may store the comparative image by mapping it with the feature information. Alternatively, the comparison target data representing the comparison target image may be stored in the database. The data to be compared may be, for example, data comparable to at least one tire image data among letters, numbers and symbols located on the side of the received tire image.

The database 2220 of the present embodiment may be constructed together with the learning database of the tire recognition module 104 discussed above or may be separately constructed.

The comparator 2230 compares the received tire image with the comparison image extracted from the database 2220 to determine whether the tire image is damaged, such as cracks or tears. That is, the comparator 2230 does not exist in the comparison target image but grasps an area existing only in the received tire image. In order to compare the received tire image with the comparison target image, it is preferable that the sizes of the two images or the positions of the minutiae points (for example, the positions of the character areas and the like) coincide with each other. To this end, the comparator 2230 adjusts the size and the direction of the received tire image or the comparison target image so that the positions of the respective minutiae coincide with each other and the sizes and shapes of the two images coincide with each other.

The damage detection unit 2240 determines the presence or absence of damage based on a different area (e.g., a line component such as a crack) recognized by the comparison unit 2230 or the like. For example, if the size of the line component detected by the comparison unit 2230 is equal to or larger than a predetermined size, the damage detection unit 2240 determines that the tire is damaged. In another example, the damage detection unit 2240 may determine whether the tire is damaged based on the number and size of line components, and then provide the terminal with information such as whether to replace the tire or the like.

23 is another block diagram of the damage recognition module 105 shown in FIG.

23, the damage recognition module 105 includes a line component detection unit 2310, a text region detection unit 2320, a damage recognition unit 2330, a DB search unit 2340, and a comparison object image database 2350 do.

The line component detection unit 2310 detects a line component in the received tire image. The linear component detector 2310 can detect only the linear component by removing the chromatic portion and the surface region in the tire image. One example of linear component detection is shown in Figs. The line component detection unit 2310 detects only line components of a black region such as a crack region, particularly cracks, among the achromatic colors in the image. The line component detection unit 2310 is not limited to the processes of FIGS. 27 to 32, Only black line components can be detected.

The character area detection unit 2320 detects a character area in the tire image. The character region detecting unit 2320 can detect a character region through detection of an area having many edges in the tire image. As another example, the character area detection unit 2320 can use the result of the character area recognized for the tire recognition in the tire recognition module 104. [

As another example, the character region detection unit 2320 can extract a character region by referring to a comparison object image extracted from the database 2340. [ That is, when the DB 23 retrieves the image to be compared with respect to the tire image after the database 2340 obtains information on the character region in advance for each image to be compared at the time of building the database 2340, 1120 can recognize the character area of the received tire image by referring to the information about the character area of the searched comparison target image.

The DB search unit 2340 searches the database 2340 for a comparison target image based on the character area of the tire image or the like.

The damage detection unit 2330 determines whether or not the line component corresponding to the character area is damaged based on the line component that is not included in the character area detected from the line component detection unit 2310.

The damage detection unit 2330 compares the image of the line component detected by the line component detection unit 2300 with the image of the comparison object searched by the DB search unit 2330 for more accurate damage detection and determines the damage. For example, if there is a portion of the line components detected by the line component detection unit 2300 that coincides with the image to be compared, this is not a damaged portion. Therefore, the damage detection unit 2330 determines whether or not the line component of the character region is removed from the line components detected by the line component detection unit 2300, .

24 to 26 are flowcharts of a tire damage recognition method according to still another embodiment.

Referring to FIG. 24, in step 2400, the server receives a tire image. In step 2410, the damage recognition module 105 recognizes the character or the like located on the side of the tire image, and grasps the tire information such as the type of the tire. In step 2420, the damage recognition module 105 searches the database based on the identified tire information and extracts a previously stored comparison image. In step 2430, the damage recognition module 105 compares the comparison image with the received tire image, and in step 2440, identifies a different area (e.g., line component) between the two images to determine whether the image is damaged. The server then provides the identified tire information and damage information to the terminal.

25, in step 2500, the server receives a tire image. In step 2510, the damage recognition module 105 detects a line component in the tire image. In step 2520, the damage recognition module 105 detects a character area in the tire image, and in step 2530, removes the line component of the detected character component in the character area. In step 2540, the damage recognition module 105 removes the line component of the character area and determines the damage based on the size and the like of the remaining line components. At this time, the damage recognition module 105 can remove the line component of the character area, compare the tire image including the remaining line components with the extracted comparison image in the database, and determine whether the damage is more accurate. A matching process between the two images can be performed to compare the comparison image with the tire image.

26, in step 2600, the server receives a tire image. In step 2610, the damage recognition module 105 removes the chromatic color from the received tire image. For example, the server may detect each component of the RGB model for each pixel of the image, and may remove the pixel whose difference between the component values of the RGB model is out of a predetermined range, thereby extracting an image consisting of only the achromatic region.

In step 2620, since the damaged area such as crack is generally black, the server extracts an area in which the brightness is equal to or less than a predetermined threshold value in the achromatic image. For example, as shown in FIG. 29, a black image including only a pixel having a brightness of less than a certain level, that is, a black region is obtained through histogram analysis of an achromatic image.

At step 2630, the damage recognition module 105 removes the surface component leaving only the line component in the black image. For example, the server uses a ridge filter to remove the area having a line shape and the area having a line shape. The server can use various image processing techniques or other kinds of filters to detect the line area.

In step 2640, the damage recognition module 105 removes the line component of the character region from the extracted line component image. For example, each character region can be extracted through the tire recognizing process, which was previously discussed. As another example, the damage recognition module 105 does not perform the entire process of the tire recognition module 104, and can extract the character region directly from the tire image by using a feature having a relatively large edge of the character region. As another example, the damage recognition module 105 can grasp the character area of the tire based on the position information of the character area of each tire stored in advance together with the comparison object image of the database.

In steps 2650 and 2660, the damage recognition module 105 removes the line component of the character area and compares the image including the remaining line components with the comparison image retrieved from the database to determine whether the image is damaged. For comparison of the image including the comparison object image and the line component, the matching of the size, shape, direction, etc. between the two images can be performed in advance.

27 and 28 are views showing an example of a method of extracting an achromatic region from a tire image.

Referring to FIG. 27, a tire in use may have various foreign substances 2710, 2720 and 2730, and these foreign substances may be various colors. On the other hand, since the damaged areas 2702, 2704 and 2706 are in the shape of a black line broken on the tire surface, the damage recognition module 105 first removes the colored areas 2720 and 2730 in the tire image 2700. For example, when a line color paint is applied to a tire, the server can remove the line shape in which the color is present from the tire image, thereby enabling a more accurate determination of the damage.

Various methods for removing a color portion in a tire image may exist, but this embodiment removes a color portion based on the sizes of R, G, and B components of the RGB model. In the case of the achromatic region, since the sizes of RGB are similar to each other, the server can remove the color portions 1520 and 1530 by removing pixels having different component sizes of RGB by a predetermined size or more. In Fig. 27, when a region in which the difference of each component of the RGB model is removed, a tire image 2800 composed of only the achromatic region is generated as shown in Fig.

FIG. 29 shows an example of a method of extracting a black region in the achromatic region of FIG. 27, and FIG. 30 is a view showing an example of an image composed of black regions obtained by applying the method of FIG.

Referring to FIG. 29, since the damaged area is close to black, the server removes a bright part (e.g., a letter part) of the achromatic color and extracts only a dark part. To do this, the server extracts only the black region by removing the pixels whose brightness is larger than a certain size by using the histogram of the brightness.

Most of the image 2800 composed of the achromatic area of FIG. 28 is removed and the damaged areas 1502, 1504, 1506, black spots 1510, There is generated an image 3000 in which the degree of the shadow part due to light remains in the area where the light is present.

31 is a diagram showing an example of a method of extracting line components in the black region of FIG.

Referring to FIG. 31, the damage recognition module 105 obtains an image 3100 from which a surface-shaped speckle 2710 or the like is removed from the black area image 3000 of FIG. Since the black speckle portion 2710 is in the form of a face, the server performs an image processing process to remove the plane component while leaving only the line component. To this end, the server may use a ridge filter that can pass line components and remove surface components. It is needless to say that various other image processing techniques can be applied to detect only the line component.

FIG. 32 is a diagram showing an example of a method of removing a line component of a character region in the line component image of FIG. 31;

Referring to FIG. 32, a line component 3210 may be included in an image 3200 in which a surface component is removed and composed only of line components. Therefore, the server performs the process of removing the line component by the character.

For example, the server grasps the start and end points of each line component and removes these line components (3210) if both the start and end points are within the text area. The character area can be grasped by various methods such as the character recognition process.

33 is a diagram showing an example of a method of comparing a line component of FIG. 31 with a comparison image of a database to recognize a damage.

Referring to FIG. 33, the damage recognition module 105 compares the tire image 3310 including the line components extracted through the process of FIG. 32 with the comparison object image 3300 extracted from the database to determine whether the image is damaged.

The damage recognition module 105 first matches the rotation angle or the size of the tire image 3310 including the line component and the comparison image 3300. For example, the tire image may be elliptical due to the weight of the vehicle, etc., while the tire image previously stored in the database may be in the form of a tire as a factory image. In addition, the positions of the minutiae such as each character region in the tire image may exist at positions different from the comparison image.

The damage recognition module 105 compares the comparison image 3300 with the comparison image 3300 to match the size, shape, rotation angle, and the like of the tire image 3310 including the line component, 3310 are corrected.

Then, the server compares the matched two images 3300 and 3310 and determines whether the line components 2702, 2704 and 2706 calculated in FIG. 31 are damaged in the comparison image 3300 or not.

35 is a diagram showing a configuration of an embodiment of the wear rate measurement module 106. In Fig.

35, the wear rate measurement module 106 includes a 3D image generation unit 3510, a tread region detection unit 3520, and a wear rate measurement unit 3530. The tire tread has deep grooves for improving braking force and driving force. Since the tire tread abuts against the surface of the road, the travel distance increases. As a result, the treads 4600 and 4610 are worn as shown in FIG. 46, thereby reducing the depth of the grooves, thereby reducing the braking force and affecting safety. The driver must directly measure the depth of the tire tread to replace the tire when the depth is shallow. There is a triangle mark next to the conventional tire tread to more easily inform the tire replacement time. However, it is inconvenient for the user to individually measure the depth of the tire tread to determine the replacement timing, and some drivers may not even know how to determine whether the wear of the tire tread is enough to replace the tire. The wear measurement module 106 according to the embodiment allows the user to easily measure tire wear on the basis of a moving image of a tire tread portion taken by a user via a camera or the like. The user can easily grasp tire wear by photographing a tire moving image through a camera such as a smart phone without having to individually measure the depth of the tire tread groove. It is also possible to inform the timing of tire replacement based on the measured tire wear. Also, when the tire replacement period has come, information on the tire can be provided together.

The 3D image generating unit 3510 generates the 3D image of the video image. An image of a three-dimensional shape can be generated using the binocular parallax generated from two-dimensional images taken in different directions. Accordingly, the 3D image generating unit 3510 separates the moving image into a plurality of still images, and then generates a three-dimensional image using the binocular parallax between the still images.

More specifically, the 3D image generation unit 3510 separates the moving image for the tire tread into a plurality of still images, grasps the correspondence relation between pixels of each still image, Capturing parameters relating to angles, distances, and the like are acquired, and the depth of each pixel is grasped on the basis of the captured image parameters, thereby generating a three-dimensional image of the tread area. A method of generating a three-dimensional shape image is shown in Figs. 37 and 38. Fig.

The tread area detecting unit 3520 detects and detects the surface area and the groove area of the tread area in the three-dimensional image. For example, since the corner between the tread groove region and the surface region in the three-dimensional image is much larger than the curvature in the other regions, the tread region detecting unit 3520 detects the curvature of each pixel in the three- The edge region is detected, and the tread groove region and the surface region are distinguished based on the detected corner region.

The wear rate measuring unit 3530 measures the wear degree by grasping the depth between the tread groove area and the surface area detected by the tread area detecting unit 3520. [ For example, the wear rate measuring unit 3530 may correct the tread groove region and the surface region in the three-dimensional shape image to an approximate plane using a planar approximation algorithm, and then grasp the depth of the tread groove based on the approximate plane. As another example, since the tire wear rate can be different from each other depending on the tread groove positions, the wear rate measuring unit 3530 divides the tread groove region into a plurality of sections, and then grasps the groove depth for each section, The tire wear rate can be measured. The tire size and the actual tire size may differ from each other. In this case, it may be difficult to accurately measure the wear amount only by the depth of the tread groove obtained from the three-dimensional shape image.

For this purpose, the wear rate measuring unit 3530 corrects the tread groove depth obtained from the three-dimensional shape image to the actual tire size or grasps the tread groove depth as a ratio of the tread width in the three-dimensional shape image, Can be measured.

For example, when the tread depth depth is to be corrected to the tire size, the wear rate measuring unit 3530 measures the tread width or tread spacing of the actual tire and the tread width of the three- Correct the size of the tread groove depth by the size.

FIG. 36 is a flowchart showing an example of a method of measuring the tire wear rate according to another embodiment.

Referring to FIG. 36, in step 3600, the wear rate measurement module 106 obtains a moving picture about a tire tread portion. In step 3610, the wear rate measurement module 106 generates a three-dimensional image of the tire tread area using the tire moving image. Then, in step 3620, the wear measurement module 106 determines the depth of the tread groove from the image of the three-dimensional shape and measures the wear degree.

37 is a flowchart illustrating an example of a method of generating a three-dimensional image from a moving image for tire wear measurement according to another embodiment, and FIG. 38 is a flowchart illustrating an example of a method of generating a three- Into spatial coordinates of the image plane.

Referring to FIG. 37, in step 3700, the wear rate measurement module 106 divides a moving image into a plurality of still images. In step 3710, the wear measurement module 106 grasps the pixel-by-pixel correspondence between the plurality of still images. For example, referring to Figure 38, an object for each pixel of a particular position of an image generated by the terminal 110 taken at different locations, that is, k-1 of the pixel of the second still image (3800), P _j, If _k-1 is the pixel P _{j, k +} ¹ of the k-th still image 3802 pixels P _{j, k,} k + 1-th still image 3804 in correspondence with each other, and stores the calculation of these correspondence relationship between each pixel . This is commonly referred to as stereo matching. In the present embodiment, various conventional methods of grasping feature points 3810 between still images and grasping the pixel-by-pixel matching relationship between still images can be applied.

In step 3720, the wear measurement module 106 calculates a relative relationship to the position of the terminal (i.e., the camera of the terminal) when each still image is captured based on the pixel-by-pixel correspondence between the plurality of still images . In other words, the wear rate measurement module 106 reversely calculates measurement parameters (camera focal length, photographing angle, camera position, etc.) at the time of photographing each still image based on the pixel-by-pixel correspondence relationship between still images.

In step 3730, the wear measurement module 106 determines the spatial coordinates of each point on the three-dimensional space based on the binocular parallax, the photographing direction, and the position of each pixel of the still images using the triangulation method or the like. And collects all the points of the spatial coordinates to generate an image on the three-dimensional space.

38, for a feature point 3810 corresponding to a k-1 still image 3800, k still image 3802, k + 1 still image 3804 for an object 3830, After capturing each of the pixels P _{k , k-1} , P _{j , k} , P _{j , k + 1} and capturing position and angle of each still image, The spatial coordinates 3820 can be obtained. By connecting the points of these spatial coordinates, a three-dimensional image is generated.

When a moving image of a tire tread is photographed, a motion occurs at a photographing position of the terminal, and a plurality of still images in the moving image obtained according to the motion of the photographing position have binocular parallax. 37 shows an example of a method of generating a three-dimensional image from a plurality of still images having binocular parallax in the moving image, and these embodiments are not limited to the method of Fig. 37, It is needless to say that an image generation method can be used.

39 is a view showing an example of an image of a three-dimensional shape acquired from a tire tread moving picture.

39, the wear rate measurement module 106 separates a moving image into a plurality of still images, and obtains a relative position and a photographing angle of the plurality of still images with respect to the plurality of still images, It is possible to obtain a plurality of still images having binocular parallax as captured.

The wear measurement module 106 generates a three-dimensional shaped image 3910 for the tire dread area based on a plurality of still images with binocular parallax.

FIG. 40 is a flowchart showing an example of a concrete method of measuring tire tread wear from an image of the generated three-dimensional shape.

Referring to FIG. 40, in step 4000, the wear rate measurement module 106 obtains an image of a three-dimensional shape as shown in FIG. 39, and first analyzes the curvature of each pixel in the image of the three-dimensional shape. In step 4010, the wear rate measurement module 106 connects pixels within a certain range of curvature magnitude similarities and mutual distances. FIG. 41 shows an example in which each area classified by the curvature size of each pixel is color-coded and displayed. The range of the curvature size for distinguishing each region can be variously set according to the embodiment.

For example, when connecting pixels having a curvature size larger than a preset threshold value, an edge region 4110 between the surface region and the groove region is detected in a three-dimensional image. In addition, when the pixels having a curvature close to zero are connected, a planar surface region and a groove region are detected. However, if each region is divided using the curvature magnitude of each pixel of the three-dimensional shape image, small noise regions may be generated as shown in FIG. Since the size of the noise regions is relatively smaller than the size of the surface region, the groove region, and the edge region, the noise region can be removed by removing regions having a predetermined size or smaller. In other words, as shown in FIG. 41, it is possible to absorb all of the regions of a predetermined size or less appearing in the surface region into a large region to form a uniform plane.

In step 4020, the wear rate measurement module 106 distinguishes the tread groove area 4120 and the surface area 4100 with respect to the area with the largest curvature, i.e., the edge area 4110. In step 4030, the wear measurement module 106 may directly determine the tire wear rate based on the depth of the tread groove area 4120, but may consider that the tread wear rate may differ from measurement position to tread groove area 4120 Thereby dividing the tread groove region into a plurality of sections.

There can be various methods for dividing the tread groove region into a plurality of sections. For example, referring to FIG. 42, it can be divided into a plurality of sections based on the direction and width of the tread groove region. Specifically, the wear measurement device identifies the intermediate axes 4200, 4210, and 4220 with respect to the direction of the tread groove region 4120 and detects the direction vectors 4230, 4240, 4250 perpendicular to the intermediate axes 4200, The width of each tread groove is determined. In addition, the wear measuring apparatus can divide the tread groove region into a plurality of sections (4260, 4262, 4264, 4266, 4268, 4270, 4272) on the basis of the direction and width.

For example, if we look at the width in the direction perpendicular to the intermediate axis with respect to the tread groove portion in the left vertical direction in FIG. 42, since the area in the middle portion is relatively large, three portions (4260, 4262, ). The remaining areas can be classified by each area based on the width. It is also possible to divide the tread groove region by various other methods such as dividing the outer tread groove region 4120 by a certain area or a predetermined length unit.

The wear measurement module 106 may correct the tread groove area 4120 and the surface area 4100 to an approximate plane. For example, referring to FIG. 45, an abrasion measuring apparatus can make a surface of a surface region and a groove region into an approximate plane 4500 by using a planar approximation algorithm such as Least Squares Fitting.

At step 4050, the wear rate measurement module 106 determines the depth of the groove area from the top surface area. For example, if the surface area of the three-dimensional shape image 4300 is separated from the tread groove area as shown in FIG. 43, the depth of the groove from the surface can be grasped as shown in FIG. Or the depth of the groove can be grasped by grasping the length of the edge area 4110 that distinguishes the surface area from the tread groove area. If the approximate plane 4500 is found as shown in Fig. 45, the depth of the groove region can be grasped based on the approximate plane. If the tread groove region is divided into a plurality of regions 4260 to 4272 as shown in FIG. 42, the depth of the groove region may be determined for each region, and then the depth of the tire tread may be the deepest region.

In step 4060, the wear measurement module 106 determines the degree of wear of the tire based on the depth of the tread groove area, and then calculates information on whether or not the tire is to be replaced or replaced, and provides the information to the terminal. Since the tire size in the image of the three-dimensional shape and the actual tire size are in a constant proportional relationship, the wear measuring apparatus can measure tire wear by correcting the actual tire size instead of using the groove depth obtained from the image of the three- have.

The present invention can also be embodied as computer-readable codes on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the computer-readable recording medium include various types of ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like. The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

The present invention has been described with reference to the preferred embodiments. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

Claims (20)

A method for providing tire information using a portable terminal,
Receiving a selection signal for an image search among search menus provided by a tire information providing system;
Driving a camera module of the portable terminal in response to the received selection signal;
Capturing a predetermined tire to obtain a tire image;
Analyzing the obtained tire image; And
And providing the tire information according to the analysis result,
Wherein the analyzing step comprises:
Converting a circular tire portion in the tire image into a linear image; Extracting a recognition target character region in the date image; And recognizing the learning character most similar to the extracted recognition target character area as the character of the recognition target character area in the group of the learned learning characters,
Wherein the step of converting into the date-
Extracting a circular tire portion in the image and rearranging straight images corresponding to a straight line perpendicular to the circumference of the circle or ellipse corresponding to the wheel region in the tire portion in the same direction to obtain the straight tire image,
Extracting a circular tire portion in the image, removing the wheel portion from the circular tire portion, converting the tire portion from which the wheel portion is removed into the linear image,
Extracting a certain region including the tire portion based on a guide line provided at the time of image capturing, and converting the extracted region into a linear image. The method according to claim 1,
The tire information includes:
Wherein the tire information includes at least one of tire product information related to tires identical or similar to the tire, damage information of the tire, and wear resistance information of the tire. 3. The method of claim 2,
The tire product information may include:
Wherein the information includes at least one of a manufacturer, a product name, a size, a product function, a manufacturing date, an offline price, an online price, price comparison information, factory price, replacement determination information, review information, Information delivery method. delete delete The method according to claim 1,
Wherein the step of converting into the date-
Converting the extracted circular tire portion into a linear image based on a guideline provided at the time of shooting the tire image;
Calculating a boundary of the wheel region based on the intensity of the image of the first date format; And
Extracting a tire portion of a circular shape in the tire image based on the boundary, and performing a second transformation on the tire image into a linear shape image. The method according to claim 6,
Wherein the step of calculating the boundaries of the wheel regions comprises:
Assigning a weight to each line of the image in the form of a direction; And
Obtaining a boundary line of each area based on intensities in the date image; And
And calculating a shortest path connecting both ends of the path image by applying a weight according to the direction to the path image formed by the boundary lines of the respective regions. 8. The method of claim 7,
Wherein the step of performing the second-
Converting the shortest path to a circular image;
Correcting the shortest path in the circular image to a circle or an ellipse;
Extracting the tire portion in the image based on the corrected circle or ellipse; And
And converting the re-extracted tire portion into a date-type timer image. The method according to claim 1,
Wherein the step of extracting the character region comprises:
And extracting each character region using at least one of a texture, a pattern, and an edge in the image in the form of a date. The method according to claim 1,
Wherein the step of extracting the character region comprises:
Applying an entropy filter to the linear image;
Acquiring a binary image composed of regions of black and white using an adaptive threshold value for an image obtained by applying the entropy filter;
Removing noise regions of a predetermined size or less from the binary image; And
And extracting the recognition target character region in the image from which the noise region has been removed. The method as claimed in claim 1,
Acquiring feature information by applying at least one filter to the recognition target character region; And
The learning character group is compared with the characteristic information of the character region to be recognized by applying the at least one filter to the learning character group to determine the learning character having the most similar character information as the character region included in the recognition target character region And recognizing the tire information as a character. The method according to claim 1,
After the camera module is driven,
Displaying at least one guide line on a screen of the portable terminal;
And displaying the tire image obtained through the camera module on the screen,
Wherein the at least one guide line comprises:
A first guide line having a circular shape and a circular second guide line having a smaller diameter than the first guide line in the first guide line,
Wherein the shape of the guide line is adaptively modified according to a tilt of the portable terminal or a distance between the camera module and the tire. The method according to claim 1,
Wherein the analyzing step comprises:
Searching for a comparison target image or data to be compared corresponding to the comparison target image in the database using at least one of the characters, numbers, and symbols located on the side of the tire image; And
Further comprising determining whether the tire image is damaged based on a difference between the tire image and the comparison object or a difference between the tire image data and the comparison object data. 14. The method of claim 13,
Wherein the step of searching for the comparison-
Converting a circular tire portion in the tire image into a linear image;
Extracting a recognition target character region in the date image;
Recognizing the learning character most similar to the extracted recognition target character region as the character of the recognition target character region in the pre-established learning character group; And
And searching the comparison image in the database based on the identified character. 14. The method of claim 13,
Wherein the analyzing step comprises:
Detecting a line component in the tire image;
Detecting a character region in the tire image;
Removing line components located in the character region from the detected line components; And
Further comprising the step of determining whether the tire is damaged based on the remaining line component after the removing step. The method according to claim 1,
The tire image is a moving image,
Generating an image of a three-dimensional shape of the tire based on the moving image; And
And measuring wear of the tire tread based on the depth of the tread area in the image of the three-dimensional shape. 17. The method of claim 16,
Wherein the step of generating the three-
Separating the moving image into a plurality of still image units;
Recognizing a pixel-by-pixel correspondence relationship between the plurality of still images;
Determining a parameter including an angle and a distance at the time of photographing the moving picture through the pixel-by-pixel correspondence relationship; And
And generating a three-dimensional shape image for the tire by grasping depth information between the plurality of still images based on the parameter. 17. The method of claim 16,
Wherein the step of measuring the wear rate comprises:
Detecting a tread groove region and a surface region based on a curvature analysis of the image of the three-dimensional shape;
Determining a depth of the groove region from the detected surface region;
And recognizing wear of the tire based on the detected depth. 19. The method of claim 18,
Wherein the step of detecting the tread groove region and the surface region comprises:
Analyzing a curvature of each pixel of the three-dimensional shape image;
Determining a region having a largest curvature among regions generated by connecting pixels having similar curvatures within a predetermined range and having mutual distances within a predetermined distance range; And
And detecting a tread groove region and a surface region separated on the basis of the region having the largest curvature. A recording medium storing a program for causing a computer to execute the method according to any one of claims 1 to 3 and 6 to 19.

Images