KR102593769B1 - Wireless Object Recognition System by Artificial Intelligence And Object Recognition Method by the Same - Google Patents

Abstract

The present invention is a wireless object recognition system using artificial intelligence that is easy to install and does not require an expensive router, in which a camera that captures images and a main device that analyzes objects by artificial intelligence are connected to an R/F communication transmission and reception module. An object recognition method using this is provided.
The wireless object recognition system based on artificial intelligence of the present invention includes a wireless camera unit installed in a place where human recognition is required, captures images in real time, and transmits the captured image data through wireless communication; and a main body unit that receives image data transmitted from the wireless camera unit and detects a person by analyzing the image data using artificial intelligence.

Description

Wireless object recognition system by artificial intelligence and object recognition method using the same {Wireless Object Recognition System by Artificial Intelligence And Object Recognition Method by the Same}

The present invention relates to a wireless object recognition system using artificial intelligence and an object recognition method using the same. More specifically, the present invention relates to a wireless object recognition system using artificial intelligence and an object recognition method using the same. More specifically, a camera installed in a place where human detection is required and a main device that analyzes objects using artificial intelligence communicate through R/F communication. This relates to a wireless object recognition system using artificial intelligence that is connected to a transmitting and receiving module and is easy to install, and an object recognition method using the same.

Artificial intelligence object recognition is a technology that processes objects present in images input from a camera in real time using artificial intelligence, and is used in fields such as security surveillance systems and medical image analysis. Such real-time object recognition must ensure both accurate recognition rate and fast recognition speed.

This artificial intelligence object recognition technology is widely used in industrial fields such as forklifts and mobile vehicles. By using real-time object recognition technology, cameras can be installed at the rear or front of forklifts, mobile vehicles, etc. to detect people among objects approaching the rear or front of forklifts or mobile vehicles, thereby preventing collisions between vehicles and people. .

The most widely used artificial intelligence object recognition system with current technology is a wired human body detection system such as CCTV, where the main controller and camera device are connected by wire. In the case of cameras actually installed in the field, they had the disadvantage of having to be installed by a person using a wired extension cable up to 15 to 20 meters long, and that installation time took a lot of time. In addition, as shown in FIG. 1, the wireless object recognition system based on the prior art requires an expensive router to be connected to the camera, resulting in high installation costs and the limitation of having to use LTE or Wi-Fi for wireless communication.

Republic of Korea Patent Publication No. 10-1418872 Republic of Korea Patent Publication No. 10-1854461 Republic of Korea Patent Publication No. 10-2390813 Republic of Korea Patent Publication No. 10-2021-0066207

The present invention is intended to solve the above problems. A camera that captures images and a main device that analyzes objects using artificial intelligence are connected to an R/F communication transmission/reception module, making installation easy and eliminating the need for an expensive router. The purpose is to provide a wireless object recognition system using artificial intelligence and an object recognition method using the same.

In order to achieve the above object, a wireless object recognition system using artificial intelligence according to an embodiment of the present invention is a wireless camera that is installed in a place where human recognition is required, captures images in real time, and transmits the captured image data through wireless communication. wealth; and a main body unit that receives image data transmitted from the wireless camera unit and detects a person by analyzing the image data using artificial intelligence.

The main body includes a receiver that receives image data transmitted from the wireless camera, an artificial intelligence control unit that analyzes the image data received from the receiver using artificial intelligence to detect a person, and displays the image data analyzed by the artificial intelligence control unit. It includes a display unit that displays a person in a specific color when a person is detected by the artificial intelligence control unit, and an alarm unit that warns with a warning sound when a person is detected by the artificial intelligence control unit.

The artificial intelligence control unit is a data learning unit that learns data so that artificial intelligence can recognize people, and an object that artificial intelligence detects people from the image data received from the receiving unit based on the content learned in the data learning unit. A detection unit, an object color indicating unit for generating and transmitting a color control signal to the display unit to display the person in a specific color on the display unit when the object detection unit detects a person in the image data, and the object detection unit generating the image data It includes a warning signal generator that generates and transmits a warning sound control signal to the warning unit when a person is detected.

The wireless camera unit includes a transmitter that transmits captured image data to the main body through wireless communication, and the frequency bands of the transmitter and the receiver of the main body are 1 GHz to 4 GHz.

There may be a plurality of wireless camera units, and the display unit is capable of dividing and displaying image data from a plurality of wireless camera units on one screen.

The main body further includes a detection target change unit that can change the detection target from a person to a human body part, a moving vehicle, or a fire at the user's request.

In addition, in order to achieve the above object, an object recognition method using a wireless object recognition system based on artificial intelligence according to an embodiment of the present invention includes a first step of installing the wireless camera unit and the main body unit; A second step in which image data captured by the wireless camera unit is transmitted to the main body unit through wireless communication; A third step in which image data received from the artificial intelligence control unit of the main body is analyzed and a person is detected; and a fourth step where, when a person is detected, the person is displayed in a specific color on the display unit and simultaneously notified with a warning sound through the alarm unit.

As described above, the present invention has the advantage of being easy to install because a camera that captures images and a main device that analyzes objects using artificial intelligence are connected to an R/F communication transmission/reception module, eliminating the need for a wired extension cable.

Additionally, the present invention has the advantage of not using an expensive router because it uses an R/F communication transmission/reception module.

Figure 1 shows a wireless object recognition system based on conventional technology.
Figure 2 is a configuration diagram of a wireless object recognition system using artificial intelligence according to an embodiment of the present invention.
Figure 3 is a configuration diagram of a wireless camera unit according to an embodiment of the present invention.
Figure 4 is a configuration diagram of the main body according to an embodiment of the present invention.
Figure 5 is a configuration diagram of an artificial intelligence control unit according to an embodiment of the present invention.
Figure 6 is a flowchart of an object recognition method using a wireless object recognition system based on artificial intelligence according to an embodiment of the present invention.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention. While describing each drawing, similar reference numerals are used for similar components.

Terms such as first, second, A, B, etc. may be used to describe various components, but the components should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

For example, a first component may be named a second component, and similarly, the second component may also be named a first component without departing from the scope of the present invention. The term “and/or” includes any of a plurality of related stated items or a combination of a plurality of related stated items.

When a component is said to be "connected" or "connected" to another component, it is understood that it may be directly connected to or connected to the other component, but that other components may exist in between. It should be. On the other hand, when it is mentioned that a component is “directly connected” or “directly connected” to another component, it should be understood that there are no other components in between.

The terms used in this application are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless explicitly defined in the present application, should not be interpreted in an ideal or excessively formal sense. No.

Hereinafter, a wireless object recognition system 10 using artificial intelligence according to an embodiment of the present invention will be described with reference to the drawings. Figure 2 is a configuration diagram of a wireless object recognition system 10 using artificial intelligence according to an embodiment of the present invention. Referring to FIG. 2, a wireless object recognition system 10 using artificial intelligence according to an embodiment of the present invention may be configured to include a plurality of wireless camera units 100 and a main body unit 200.

First, a plurality of wireless camera units 100 are installed in places where human recognition is required, capture images in real time, and transmit the captured image data through wireless communication. The plurality of wireless camera units 100 of the present invention can be installed on the front or rear of a mobile vehicle, such as a forklift or mobile vehicle, that has a risk of collision with a person at an industrial site. In order to photograph objects approaching from various angles, the present invention may be provided with a plurality of wireless camera units 100. Figure 7 shows a schematic diagram of a wireless camera unit installed in a mobile vehicle. As shown in FIG. 7, a plurality of wireless camera units 100 installed in places where human recognition is required include a wireless camera 100A for the front left, a wireless camera 100A for the front left, and a wireless camera 100A for the front right of the mobile vehicle 500. Including the wireless camera for the front right (100B), the wireless camera for the rear left (100C) installed on the rear left, and the wireless camera for the rear right (100D) installed on the rear right, it is possible to capture images of the front, rear, and left and right sides of the mobile vehicle. You can.

Figure 3 is a configuration diagram of a plurality of wireless camera units 100 according to an embodiment of the present invention. Referring to FIG. 3, the plurality of wireless camera units 100 according to an embodiment of the present invention includes a plurality of cameras 110 and a transmitter 120, and may additionally include an external speaker.

The camera 110 captures images in real time at a location where a plurality of wireless camera units 100 are installed. In general, the camera 110 is classified according to the sensor used in the camera 110, and includes a CCD sensor type and a CMOS sensor type. Both CCD sensors and CMOS sensors transmit light brightness signals (charge) to the processor, but depending on the design of the image sensor, there is a method that transmits the charge directly (CCD) or a method that transmits the charge of each pixel as a digital signal (CMOS). It can be divided into Charge loss may occur during conversion to digital signals, so CCD is relatively superior in terms of image quality, but CMOS is becoming more integrated based on semiconductor technology, and for this reason, it has a small size and excellent power efficiency. there is. In the present invention, either the CCD sensor type or the CMOS sensor type can be selected and used, but it is preferable to use the CMOS sensor type camera 110, which consumes less power and enables high-speed processing.

The transmitter 120 transmits image data captured by the camera 110 to the receiver 210 of the main body 200 through wireless communication. The transmitter 120 and the receiver 210 of the present invention can use RF communication as wireless communication, and can use an RF module driven by RF communication as a device. RF communication is communication that operates at radio frequency, and the communication frequency band is generally 300kHz to 300GHz. RF communication can transmit signals farther than infrared communication, and communication is possible even if there are obstacles between the transmitter 120 and the receiver 210. Additionally, RF communication uses a specific frequency band, so the impact of interference is low. There is an advantage. An RF module (radio-frequency module) is a small electronic device used to transmit and receive wireless signals between two devices, that is, between the transmitter 120 and the receiver 210. Several carrier frequencies are commonly used in commercial RF modules, including frequencies in the ISM radio band for industrial, scientific, and medical applications, such as 433.92 MHz, 915 MHz, and 2400 MHz. These frequencies are used based on national and international regulations regarding the use of radios for communications. Near-field devices may also use license-exempt frequencies such as 315 MHz and 868 MHz. In the present invention, the frequency band of the transmitter 120 and the receiver 210 may be 1 to 4 GHz, and it is preferable to use 2.5 GHz, which is a stable frequency band, in order to stably transmit and receive video data.

Next, the main body unit 200 receives image data transmitted from the plurality of wireless camera units 100 and analyzes the image data using artificial intelligence to detect people. If a plurality of wireless camera units 100 are installed on the front or rear of a mobile vehicle, such as a forklift or mobile vehicle, that has a risk of collision with people at an industrial site, the main body unit 200 is installed on the dashboard or the dashboard of the forklift or mobile vehicle. It can be installed nearby. Drivers of forklifts, mobile vehicles, etc. can check objects approaching the rear or front in real time through the display unit 230 provided in the main body 200. The main body 200 according to an embodiment of the present invention may use an embedded Linux system, and the artificial intelligence may be a program written in Python. Additionally, the result of detecting a person can be displayed through the alarm unit 240 or display unit 230 connected to the GPIO port of the embedded system. The output value of the GPIO port is transmitted to another MCU module, which has the advantage of improving response speed by activating the interrupt function. Figure 4 is a configuration diagram of the main body 200 according to an embodiment of the present invention. Referring to FIG. 4, the main body 200 according to an embodiment of the present invention includes a receiving unit 210, an artificial intelligence control unit 220, a display unit 230, an alarm unit 240, and an internal speaker 240C. It can be configured as follows.

The receiver 210 receives image data transmitted from the wireless camera 110. The receiver 210 can use RF communication as wireless communication like the transmitter 120, and can use an RF module driven by RF communication as a device.

The artificial intelligence control unit 220 analyzes the image data received from the receiver 210 using artificial intelligence to detect people. Figure 5 is a configuration diagram of the artificial intelligence control unit 220 according to an embodiment of the present invention. Referring to FIG. 5, the artificial intelligence control unit 220 according to an embodiment of the present invention includes a data learning unit 221, an object detection unit 222, an object color indicating unit 223, and a warning signal generating unit 224. It can be configured as follows. The object recognition of the present invention uses an object detection model based on artificial intelligence, and the object detection model can largely use One-Stage Detector and Two-Stage Detector. Two types of models use Classification and Region Proposal. Classification is an algorithm that classifies what kind of object a specific object is, and Region Proposal is an algorithm that quickly finds an area where an object is likely to be. One-stage detector is a method of obtaining results by simultaneously performing classification and regional proposal. Additionally, the Two-stage Detector is a method of obtaining results by sequentially performing Classification and Regional Proposal. One-stage detectors include YOLO, SSD, and RetinaNet, and two-stage detectors include R-CNN, Fast R-CNN, and Faster R-CNN. As the artificial intelligence algorithm of the present invention, at least one of YOLO, SSD, RetinaNet, R-CNN, Fast R-CNN, and Faster R-CNN can be selected and used.

The data learning unit 221 learns data so that artificial intelligence can recognize people from image data. The data learning unit 221 of the present invention has already repeatedly learned labeled data using machine learning or deep learning.

The object detection unit 222 detects people using artificial intelligence from the image data received from the receiving unit 210 based on content learned in the data learning unit 221. The object detection unit 222 analyzes image data captured by a plurality of wireless camera units 100 in real time based on data learned in data learning, and detects people among the image data using artificial intelligence. When a person is detected in the image data, the object detection unit 222 generates a signal to notify the object color indicator unit 223 and the warning signal generator 224, and sends it to the object color indicator unit 223 and the warning signal generator 224. It may be provided with a detection signal transmitting unit that transmits to the generating unit 224. When a person is detected in the object detection unit 222, the detection signal transmission unit generates and transmits a person detection signal to the object color indicator unit 223. When a person is no longer detected in the object detection unit 222, the detection signal transmission unit generates and transmits a person detection stop signal to the object color indicator unit 223. The object color indicating unit 223, which has received the human detection stop signal, does not transmit a color control signal to the display unit 230. Additionally, when a person is detected in the object detection unit 222, the detection signal transmission unit generates and transmits a person detection signal to the warning signal generation unit 224. When a person is no longer detected in the object detection unit 222, the detection signal transmission unit generates and transmits a person detection stop signal to the warning signal generation unit 224. The warning signal generator 224, which has received the person detection stop signal, no longer transmits the warning sound control signal to the warning unit 240.

When the object detection unit 222 detects a person in the image data, the object color indicating unit 223 generates and transmits a color control signal to the display unit 230 so that the person is displayed in a specific color on the display unit 230. If a person is detected in the image data by artificial intelligence in the object detection unit 222, the detection signal transmission unit of the object detection unit 222 generates a person detection signal and transmits it to the object color indicator unit 223. The object color indicating unit 223 may receive a person detection signal from the object detection unit 222 and change the color of the recognized person or display a line on the outside of the person image so that the person can be distinguished from the image data. For example, when the object color indicating unit 223 generates a color control signal that displays the outline of a person detected in the image data as a red dotted line and transmits it to the display unit 230, the display unit 230 displays the outline of the person. The image processed with red dotted lines is displayed.

In addition, the object color indicator 223 includes a person classification color indicator, and the person classification color indicator distinguishes people pre-stored in the data learning unit 221 from people not pre-stored in different colors and displays the information on the display unit 230. It can be expressed. For example, if a pre-recorded person appears in a captured video, the outline of the person is displayed with a blue dotted line. If a person who has not been pre-recorded appears in the captured video, the outline of the person is displayed with a red dotted line. It can display images. Accordingly, the driver can determine whether the person appearing in the video is a person working in the workplace, or a person coming from outside or working in another workplace, and can vary the level of vigilance.

The warning signal generation unit 224 generates and transmits a warning sound control signal to the warning unit 240 when the object detection unit 222 detects a person in the image data. If a person is detected in the image data by artificial intelligence in the object detection unit 222, the detection signal transmission unit of the object detection unit 222 generates a person detection signal and transmits it to the warning signal generation unit 224. The warning signal generator 224 receives a person detection signal from the object detector 222, generates a warning sound control signal and transmits it to the alarm unit 240 so that the alarm unit 240 can emit a warning sound.

In addition, the warning signal generator 224 includes a person-discriminating warning sound generating unit, and the person-discriminating warning sound generating unit can generate warning sounds differently by distinguishing between people pre-stored in the data learning unit 221 and people not pre-stored. For example, if a pre-recorded person appears in the captured video, an alarm sound is generated along with the name of the pre-recorded person indicating that they have entered the dangerous area. If a person not pre-recorded appears in the recorded video, a random person enters the dangerous area. An alarm sound can be generated to indicate entry. Therefore, the driver can tell without going through the display whether the person appearing in the video is a person working in the workplace, or a person coming from outside or working in another workplace, and can vary the level of vigilance.

The display unit 230 displays image data analyzed by the artificial intelligence control unit 220, and when a person is detected by the artificial intelligence control unit 220, the person is displayed in a specific color. Data captured by the plurality of wireless camera units 100 is transmitted to the main body unit 200 and analyzed in real time by the artificial intelligence control unit 220, and the image data analyzed in real time is displayed through the display unit 230. . Since the wireless object recognition system 10 based on artificial intelligence of the present invention is installed on forklifts, mobile vehicles, etc., the analysis by artificial intelligence and display on the display unit 230 at the time when the image is captured must be done almost simultaneously and there is a time difference. should not occur significantly. Accordingly, the display unit 230 displays image data captured by the plurality of wireless camera units 100 at almost the same time through the artificial intelligence control unit 220. Additionally, when a person is detected in the image data by the object detection unit 222, the display unit 230 may receive a color control signal from the object color indicator unit 223 and display the person in a specific color so that the person can be distinguished. The display unit 230 of the present invention can use various types of displays, for example, CRT (Cathode Ray Tube), PDP (Plasma Display Panel), OLED (Organic LED), LED (Light Emitting Diode), FED ( Field, Emission Display (VFD), Vacuum Fluorescent Display (VFD), Liquid Crystal Display (LCD), Electrophoretic Display (EPD), and Electrochromic Display (ECD), but not limited to displays not listed here. Can be used including.

There may be a plurality of wireless camera units 100 of the present invention, and the display unit 230 may divide the image data of the plurality of wireless camera units 100 into multiple pieces and display them on one screen. For example, if four wireless camera units 100 are installed, the display unit 230 may be divided into four screens and displayed. That is, the screen of the display unit 230 can be divided according to the number of wireless cameras 110. Since the display unit 230 must display images captured by the wireless cameras 110 on the screen, the screen is divided by the number of wireless cameras 110 and images captured by a plurality of wireless cameras 110 are displayed simultaneously on the screen. It can be done.

The display unit 230 of the present invention may include a screen splitter. The screen dividing unit includes a first image from a wireless camera for the front left side provided on the front left side of the mobile vehicle, a second image from a wireless camera provided for the front right side provided on the front right side, and a wireless camera provided for the rear left side provided on the rear left side of the mobile vehicle. The third image and the fourth image from the wireless camera for the rear right side can be displayed in four areas divided into equal areas on one screen. In addition, the screen dividing unit includes a screen size adjusting unit, and the screen size adjusting unit, when a person is observed in any one of the first to fourth images, changes the image in which the person is observed to the image in which the person is not observed. It can be displayed by enlarging it larger than the image. For example, when a person is observed in the first image, the size of the first image may be increased by 1.5 to 2 times, and the size of the second to fourth images may be reduced by 0.3 to 0.5 times. In addition, the screen dividing unit includes a screen position adjusting unit, and the screen position adjusting unit adjusts the image in which the person is observed to the center of the screen when a person is observed in any one of the first to fourth images. It can be positioned. Therefore, the driver can continue driving while observing more closely without manually enlarging the image of the dangerous area.

The alarm unit 240 warns with a warning sound when a person is detected by the artificial intelligence control unit 220. If a person is detected in the image data by artificial intelligence in the object detection unit 222, the detection signal transmission unit of the object detection unit 222 generates a person detection signal and transmits it to the warning signal generation unit 224. The warning signal generating unit 224 generates a warning sound control signal and transmits it to the warning unit 240. The alarm unit 240 receives a warning sound control signal and notifies the user through a warning sound that a person has been detected. The main body 200 of the present invention further includes an inner speaker, and the alarm unit 240 can transmit a warning signal to the inner speaker. When a person is detected, an alarm or warning sound may be used to inform the driver of the mobile vehicle that a person has been detected in the front or rear of the wireless camera unit 100 where the plurality of wireless camera units 100 are installed.

Meanwhile, as shown in FIGS. 8(a) and 8(b), a plurality of external speakers may be provided on the sides of the plurality of wireless camera units 100 of the mobile vehicle 500. The external speakers are the front left speaker (240A) provided on the front left (Ⅰ), the front right speaker (240B) provided on the front right (Ⅱ), and the rear speaker (240B) provided on the rear left (Ⅲ), based on the mobile vehicle. Including the left speaker (240C) and the rear right speaker (240D) provided on the rear right side (IV), warning broadcasts can be made from the front, rear, and left and right sides of the mobile vehicle. The warning unit 240 can warn more efficiently by selectively transmitting a warning signal only to the inner speaker 240C provided in the main body and the outer speaker 240A close to the wireless camera unit 100A where the person 300 is photographed. You can.

The alarm unit 240 may include a location specification unit that transmits a warning signal only to the inner speaker attached to the main body 200 and the outer speaker installed in the specific wireless camera unit from the image of the specific wireless camera unit in which the person is detected. The location specification unit may include an internal location specification unit and an external location specification unit. In the present invention, since a plurality of wireless camera units 100 are installed, the display unit 230 is divided into multiple screens. In general, four to six wireless camera units 100 may be installed, and when a person is detected by a specific plurality of wireless camera units 100, the user may use the plurality of wireless camera units 100 installed at a certain location. ) may not be recognized. In the present invention, an internal location specification unit is provided to inform the driver of a mobile vehicle of the specific location of the wireless camera by informing the driver of the mobile vehicle of the location of the person in the wireless camera unit 100 among the plurality of wireless camera units 100 in which the person was detected through the main body speaker. A warning broadcast is selectively made, and an external location specification unit is provided so that a warning broadcast can be made to a person located outside the mobile vehicle by selecting a specific external speaker.

Meanwhile, as shown in Figure 8 (c), the external location specification unit provides warning directions of the mobile vehicle to the person according to the direction in which the person is located relative to the mobile vehicle, such as: There is a vehicle on the right side in front, There is a vehicle on the left side in front, Warnings that there is a vehicle on the right side of the rear or that there is a vehicle on the left side of the rear can be selectively broadcast. The artificial intelligence control unit analyzes the received video data to detect the direction in which the person is heading with respect to the mobile vehicle, and a warning broadcast can be broadcast from the speaker according to the detected direction. For example, the front of the mobile vehicle is defined as north, the right side of the mobile vehicle is defined as east, the left side of the mobile vehicle is defined as west, and the rear of the mobile vehicle is defined as south, with the mobile vehicle facing north and the mobile vehicle facing north. When there is a person on the front left of the car and the person is facing north (①), the front left speaker broadcasts a warning that “there is a vehicle on the rear right”, and when the person is facing east (②) ), the front left speaker broadcasts a warning message saying "There is a vehicle on the right front," and when a person is facing west (③), the front left speaker broadcasts a warning message saying "There is a vehicle on the rear left side." And, if the person is facing south (④), the front left speaker broadcasts a warning that “there is a vehicle on the left front,” and artificial intelligence analyzes and informs the location of the mobile vehicle according to the person’s direction. This allows people to evacuate effectively.

The location specification unit stores broadcast content about the location where each of the plurality of wireless camera units 100 is installed. If a person is detected in the image data by artificial intelligence in the object detection unit 222, the detection signal transmission unit of the object detection unit 222 generates information on the plurality of wireless camera units 100 in which the person was detected and a person detection signal to warn the user. It is transmitted to the signal generator 224. The warning signal generator 224 transmits the warning sound control signal and the information of the plurality of wireless camera units 100 to the location specification unit of the alarm unit 240, and the location specification unit detects a specific person in the plurality of wireless camera units 100. Stored broadcast content related to the location can be played to inform the user.

The main body 200 includes a detection target changer that can change the detection target from a person to a human body part, a moving vehicle, or a fire at the user's request. The user can change the detection target through the detection target change unit. Check the detection target list on the settings screen of the display unit 230 through the input device provided in the main body 200 and select the desired detection target from the detection target list. You can choose. For example, the detection target list can be objects that can be detected by artificial intelligence and include people, human hands, human body parts, cars, fire, etc. The detection target included in the detection target list must match the detection target already learned by the data learning unit 221 of the artificial intelligence control unit 220, and when the detection target change unit changes the detection target, the object detection unit 222 detects the changed object. Detect objects according to the target.

Hereinafter, an object recognition method using the artificial intelligence wireless object recognition system 10 according to an embodiment of the present invention will be described with reference to the drawings. Figure 6 is a flowchart of an object recognition method using a wireless object recognition system 10 based on artificial intelligence according to an embodiment of the present invention. Referring to FIG. 6, an object recognition method using the artificial intelligence wireless object recognition system 10 according to an embodiment of the present invention may be configured to include the following four steps.

First step (S10): Step in which a plurality of wireless camera units 100 and main body units 200 are installed.

Second step (S20): A step in which image data captured by a plurality of wireless camera units 100 is transmitted to the main body unit 200 through wireless communication.

Third step (S30): A step in which image data received from the artificial intelligence control unit 220 of the main body 200 is analyzed and a person is detected.

Fourth step (S40): When a person is detected, the person is displayed in a specific color on the display unit 230 and simultaneously notified with a warning sound through the alarm unit 240.

The first step (S10) is a step of installing the wireless object recognition system 10 based on artificial intelligence in a place where object recognition is required. The wireless object recognition system 10 based on artificial intelligence of the present invention is installed on forklifts and mobile vehicles at industrial sites. A plurality of wireless camera units 100 are installed on the front and rear of the forklifts and mobile vehicles, and The main body 200 is installed near the instrument panel of the vehicle. The present invention has the advantage of eliminating the inconvenience of installing long separate extension cables because the plurality of wireless camera units 100 and the main body unit 200 are connected through RF communication.

In the second step (S20), after the plurality of wireless camera units 100 and the main body unit 200 are installed, the plurality of wireless camera units 100 operate and the captured image data is transmitted to the main body unit 200 through wireless communication. This is the stage where it is transferred to . A plurality of wireless camera units 100 are installed on the front or rear of a forklift or mobile vehicle to start capturing images, and the captured image data is transmitted to the main body 200 through RF communication. The plurality of wireless camera units 100 are equipped with a transmitter 120, and the main body 200 is equipped with a reception unit 210, so that image data is transmitted to the main body 200 through an RF module.

The third step (S30) is a step of detecting a person by analyzing image data received from the artificial intelligence control unit 220 of the main body 200. Image data received through the receiver 210 is detected by recognizing a person through the object detection unit 222 based on content learned in the data learning unit 221.

The fourth step (S40) is a step where, when a person is detected by the object detection unit 222, the detected person is displayed on the display unit 230 in a specific color and simultaneously notified with a warning sound through the alarm unit 240. When a person is detected in the object detection unit 222, the detection signal transmission unit generates and transmits a person detection signal to the object color indicator unit 223. The object color indicating unit 223 that receives the person detection signal generates and transmits a color control signal to the display unit 230. People are displayed in a specific color on the display unit 230, and the user can distinguish people from other objects. Additionally, when a person is detected in the object detection unit 222, the detection signal transmission unit generates and transmits a person detection signal to the warning signal generation unit 224. The warning signal generation unit 224, which receives the person detection signal, generates a warning sound control signal and transmits it to the warning unit 240. The alarm unit 240 displays the detected person in a specific color on the display unit 230 and sounds an alarm at the same time to warn the user that the person has been detected. If a person is not detected in the image data transmitted from the plurality of wireless camera units 100, the display unit 230 no longer displays in a specific color and the alarm unit 240 ends playing the alarm sound.

The present invention is not limited to the above-mentioned embodiments, but can be manufactured in various different forms, and those skilled in the art may manufacture the present invention in other specific forms without changing the technical idea or essential features of the present invention. You will understand that it can be done. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive.

10: Wireless object recognition system using artificial intelligence
100: wireless camera unit
110: camera
120: Transmitting unit
200: main body
210: receiving unit
220: Artificial intelligence control unit
221: Data learning department
222: object detection unit
223: Object color indicator
224: Warning signal generation unit
230: display unit
240: Alarm unit

Claims (7)

A wireless camera unit installed in a place where human recognition is required to capture images in real time and transmit the captured image data through wireless communication; And a main body unit that receives image data transmitted from the wireless camera unit and detects a person by analyzing the image data using artificial intelligence,
The main body includes a receiver that receives image data transmitted from the wireless camera, an artificial intelligence control unit that analyzes the image data received from the receiver using artificial intelligence to detect a person, and displays the image data analyzed by the artificial intelligence control unit. It includes a display unit that displays a person in a specific color when a person is detected by the artificial intelligence control unit, and an alarm unit that warns with a warning sound when a person is detected by the artificial intelligence control unit,
The artificial intelligence control unit is a data learning unit that learns data so that artificial intelligence can recognize people, and an object that artificial intelligence detects people from the image data received from the receiving unit based on the content learned in the data learning unit. A detection unit, an object color indicating unit for generating and transmitting a color control signal to the display unit to display the person in a specific color on the display unit when the object detection unit detects a person in the image data, and the object detection unit generating the image data When a person is detected, it includes a warning signal generator that generates and transmits a warning sound control signal to the alarm unit,
The wireless camera unit includes a transmitter that transmits captured image data to the main body through wireless communication, and the image data transmitted from the transmitter is received by a receiver that uses an R/F communication transmission/reception module in the main body, and the receiver unit The frequency band is 1GHz to 4GHz,
The wireless camera unit installed in the place where human recognition is required includes a wireless camera for the front left side of the mobile vehicle, a wireless camera for the front right side provided on the front right side, a wireless camera for the rear left side provided on the rear left side, and a rear wireless camera provided for the front left side of the mobile vehicle. It captures images from the front, back, and left and right sides of the mobile vehicle, including the wireless camera for the rear right side, which is installed on the right side.
The display unit displays the first image from the wireless camera for the front left, the second image from the wireless camera for the front right, the third image from the wireless camera for the rear left, and the fourth image from the wireless camera for the rear right on one screen. It is divided into multiple parts and displayed,
The main body further includes a detection target change unit that can change or add a detection target from a person to a human body part, a moving vehicle, or a fire at the user's request,
The object color indicator includes a person classification color indicator, and the person classification color indicator divides people pre-stored in the data learning unit and people not pre-stored in the data learning unit into different colors and displays them on the display unit. When the pre-saved person appears, the outline of the pre-saved person is displayed with a blue dotted line, and when a non-pre-saved person appears in the captured image, the outline of the non-pre-saved person is shown with a red dotted line. represents an image that has been
The warning signal generator includes a person-discriminating warning sound generating unit, and the person-discriminating warning sound generating unit generates different warning sounds by distinguishing between people pre-stored in the data learning unit and people not pre-stored,
The display unit includes a screen splitter, and the screen splitter includes a screen size adjuster, and the screen size adjuster controls the screen size adjuster to detect the person when a person is observed in any one of the first to fourth images. The displayed image is enlarged and displayed larger than the image in which no person is observed, and the screen division unit includes a screen position adjustment unit, and the screen position adjustment unit allows a person to observe any one of the first to fourth images. In this case, the image in which the person is observed is located in the center of the screen,
The main body further includes an internal speaker, and is provided with a plurality of external speakers on the sides of the plurality of wireless camera units of the mobile vehicle, wherein the plurality of external speakers include a front left speaker provided on the front left side of the mobile vehicle, It includes a front right speaker provided on the front right, a rear left speaker provided on the rear left, and a rear right speaker provided on the rear right,
The warning unit includes a location specification unit that transmits a warning signal only to the inner speaker attached to the main body and the outer speaker installed in the specific wireless camera unit from the image of the specific wireless camera unit in which the person is detected, and the location specification unit determines the internal location. It has a specification unit and an external location specification unit, wherein the internal location specification unit broadcasts a warning to inform the driver of the mobile vehicle of the specific location of the wireless camera as to which wireless camera unit among the plurality of wireless camera units has captured the person,
The external location specification unit analyzes the image data received from the artificial intelligence control unit, detects the direction in which the person is facing with respect to the mobile vehicle, and warns the direction of the mobile vehicle to the person according to the direction in which the person is located with respect to the mobile vehicle. Object recognition using a wireless object recognition system based on artificial intelligence that broadcasts a warning that there is a vehicle on the front right, a vehicle on the front left, a vehicle on the rear right, and a vehicle on the rear left. It is a method,
A first step of installing the wireless camera unit and the main body unit; A second step in which image data captured by the wireless camera unit is transmitted to the main body unit through wireless communication; A third step in which image data received from the artificial intelligence control unit of the main body is analyzed and a person is detected; and a fourth step where, when a person is detected, the person is displayed in a specific color on the display unit and simultaneously notified with a warning sound through the alarm unit. delete delete delete delete delete delete