KR102361488B1 - Surround Internet Protocol Camera - Google Patents

Abstract

Provided is a surround IP camera, which includes: a photographing unit capable of omnidirectional video and omnidirectional infrared photographing; a control unit for tracking and observing a subject in an infrared image photographed through the photographing unit, but determining that a problem has occurred in the subject based on the body temperature distribution and body temperature change amount of a non-moving subject; and a communication unit transmitting the result determined by the control unit.

Description

Surround IP Camera {Surround Internet Protocol Camera}

The present invention relates to a surveillance camera device, and more particularly, to a surround IP camera.

In general, a cctv camera (surveillance camera) is a device that transmits an image using a cable television in a specific building or a specific facility, and is widely used in various fields.

For example, users install cctv cameras in facilities such as production plants, financial institutions, train stations and subways, steelworks ships and construction sites, distributors, leisure facilities, highways and city traffic networks, and use a small number in their central control room (or security room). The personnel can monitor a wide area at a glance.

In addition, it can be operated in various ways in process supervision, warehousing, control, and automatic home office management, and it can be recorded and used as post-mortem data.

In addition, the maximum monitoring effect can be achieved with the minimum number of people, and when a criminal activity is discovered, it can be immediately eliminated.

The general configuration of such a cctv camera is that the body of the camera is rotatably installed by a pan head, and the video signal output from the camera is connected to be input to the pan head controller of the control room, and this pan head controller controls the pan head. connected

However, in this configuration, when an image signal photographed through the lens of the camera is input to the pan head controller, the user in the control room checks the image through the monitor. Therefore, the user manually adjusts the pan head in the control room to move the camera connected through the shaft up, down, left, and right to take pictures while tracking the object. In the case of such a conventional device, the situation can be sensed only when the user is checking the monitor, and active monitoring is difficult when the user (administrator) is absent.

KR 10-2010-0061890 A

The present invention has been proposed to solve the technical problem as described above, and it is possible to automatically determine that a problem has occurred in the object based on the body temperature distribution and body temperature change of the motionless object by tracking and observing the object in the infrared image. Surround IP camera is provided.

In addition, there is provided a surround IP camera capable of dividing a photographed image into a plurality of sub-regions and independently and automatically determining the presence or absence of abnormalities in each sub-region.

According to an embodiment of the present invention for solving the above problems, a photographing unit capable of omnidirectional imaging and omnidirectional infrared imaging, and tracking and observing an object in an infrared image captured through the photographing unit, but the body temperature distribution and A surround IP camera is provided, comprising: a control unit that determines that a problem has occurred in a subject based on an amount of body temperature change; and a communication unit that transmits a result determined by the control unit.

In addition, the control unit included in the present invention divides the image photographed through the photographing unit into a plurality of sub-regions and independently determines the presence or absence of abnormalities in each sub-region. It is characterized by judging the type of fire through color and fire area, and predicting risk factors by pre-projecting the type of fire on a sub-area located in the direction of fire progress.

In addition, it further includes a vibration detection unit for detecting the vibration, the control unit is characterized in that by analyzing the waveform of the vibration detected through the vibration detection unit predicts a risk factor corresponding to the vibration waveform.

In addition, according to another embodiment of the present invention, a photographing unit capable of omnidirectional imaging and omnidirectional infrared imaging, and an image captured through the photographing unit are divided into a plurality of sub-regions to independently determine the presence or absence of abnormalities in each sub-region. , a control unit that predicts and alerts risk factors by judging the fire type through the color of the flame and the fire area of the image taken in the sub-region determined as having an abnormality, and pre-projecting the fire type in the sub-region located in the direction of the fire There is provided a surround IP camera, characterized in that.

In addition, the control unit included in the present invention is characterized in that after detecting the ignition point when the fire is detected, the estimated time for the fire to arrive from the ignition point to each sub-region is calculated.

The surround IP camera according to an embodiment of the present invention tracks and observes an object in an infrared image, but can automatically determine that a problem has occurred in the object based on the body temperature distribution and body temperature change of the non-moving object.

In addition, the surround IP camera can divide the captured image into a plurality of sub-regions and automatically determine whether there is an abnormality in each sub-region independently.

1 is a block diagram of a surround IP camera 100 according to an embodiment of the present invention.
2 is an exemplary view of an omnidirectional photographed image taken by the surround IP camera 100
3 is an exemplary diagram illustrating a process of processing an image in the surround IP camera 100
4 is a block diagram of an intelligent monitoring system 1 having a surround IP camera 100
5 is an exemplary view showing an image generated by the intelligent monitoring system (1)

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings in order to describe in detail enough that a person of ordinary skill in the art to which the present invention pertains can easily implement the technical idea of the present invention.

1 is a block diagram of a surround IP camera 100 according to an embodiment of the present invention.

The surround IP camera 100 according to the present embodiment includes only a simple configuration for clearly explaining the technical idea to be proposed.

Referring to FIG. 1 , the surround IP camera 100 includes a photographing unit 110 , a vibration detection unit 120 , a sound recognition unit 130 , a LiDAR sensor 140 , and a control unit 190 .

The main operations of the surround IP camera 100 configured as described above are as follows.

The photographing unit 110 is configured to enable omnidirectional imaging and omnidirectional infrared imaging. That is, the photographing unit 110 includes a camera for photographing a visible ray region and a thermal imaging camera for photographing an infrared region.

Basically, the photographing unit 110 includes a single lens module for photographing a 360-degree direction.

That is, the photographing unit 110 refracts incident light, and has a side surface formed of a convex light refraction surface downward, an upper surface formed of a light reflection surface on which the light incident on the light refraction surface is reflected, and a lower portion of the light reflection surface. It includes a catadioptric lens having a lower surface made of a light exit surface through which light reflected from the light reflection surface is emitted downward.

In addition, the photographing unit 110 includes a catadioptric lens holder having a light transmitting part of a transparent material to surround and accommodate the side surface of the catadioptric lens, and to cover the upper surface of the catadioptric lens, and is fastened to the catadioptric lens holder A cover for fixing the catadioptric lens, at least one light transmitting lens through which light emitted downward through the light exit surface is transmitted, and an image sensor for capturing the light transmitted through the light transmitting lens, It may be manufactured to have a barrel assembly coupled to a catadioptric lens holder.

The controller 190 generates an omnidirectional flat image by image processing the captured image of the photographing unit 110 .

The controller 190 may obtain an original image photographed using an omnidirectional optical lens, and may obtain a circular image by preferentially removing unnecessary image portions. The circular image may be divided for partial image processing and partial distortion reduction to generate a flat image. A unit of the divided circular image (a circular division unit image) may be divided in consideration of resolution, arc, pixel, image distortion, and the like.

The surround IP camera 100 operates to detect a visible ray image, a thermal image, and motion information of an object of a preset photographing area. The surround IP camera 100 may apply a Time Lapse function that is recorded and stored only when there is a movement of an object.

Since the surround IP camera 100 includes a photographing unit 110 and a LiDAR/RADAR sensor 140, it is possible to effectively monitor a photographing area even at night.

That is, the photographing unit 110 captures a thermal image of the photographing area. A thermal imaging camera is a non-contact temperature measurement device. A thermal imaging camera detects infrared energy radiated, transmitted or reflected by any material at temperatures above absolute zero (0°Kelvin) and converts this energy factor into a temperature measurement or thermogram. Thermography outputs the thermal image of an object that radiates, transmits, or reflects the infrared energy displayed by the thermal imaging camera.

The thermal imaging camera transmits ultra-wide/low-power very narrow pulses and restores the received pulse signal with precise resolution. The RF transceiver chipset for impulse radar, removes clutter from the received radar signal, and collects features for moving objects (velocity) , posture estimation (gait), and the characteristics of a stationary object (respiration) can be used by applying a radar signal processing algorithm.

In addition, the lidar (LiDAR)/radar (RADAR) sensor 140 is a device capable of detecting a distance to an object, a direction, a speed, a temperature, a material distribution and concentration characteristics, etc. by shining a laser on the target.

The configuration of the lidar sensor may be divided into a laser transmitter, a laser detector, a signal collection and processing part, and a part for transmitting and receiving data. In addition, the lidar sensor may be divided into a time-of-flight (TOF) method and a phase-shift method according to a modulation method of a laser signal.

In the TOF method, it is possible to measure the distance by measuring the time that the laser emits a pulse signal and the reflected pulse signals from objects within the measurement range arrive at the receiver. The phase-shift method is a method of calculating the time and distance by emitting a laser beam that is continuously modulated with a specific frequency and measuring the amount of phase change of a signal reflected from an object within the measurement range.

As the laser light source, laser light sources having a specific wavelength or having a wavelength tunable in a wavelength range from 250 nm to 11 μm are used, and recently, a semiconductor laser diode capable of small size and low power is widely used. In particular, the wavelength of the laser directly affects the transmittance to the atmosphere, clouds, rain, etc. and eye-safety. Basically, laser power, wavelength, spectral characteristics, pulse width and shape, receiver sensitivity and dynamic range, and optical filter and lens characteristics are the main factors that determine the performance of LiDAR. In addition, Field Of View (FOV) indicating the measurement angle of the receiver, field stop to select the measurement range, and the FOV overlap characteristics of the laser beam and the receiver are also important items. The minimum time for unit data collection with respect to the speed of light is a factor that determines the range resolution, and therefore data collection and processing within a few ns is required for a distance resolution of 1 m or less.

Accordingly, the controller 190 may identify the object based on the photographing information of the LiDAR/RADAR sensor 140 . That is, the controller 190 may identify whether the sensed object recognizes a person, an animal, or an object based on the information of the LiDAR/RADAR sensor 140 .

2 is an exemplary view of an omnidirectional photographed image taken by the surround IP camera 100 .

The left image of FIG. 2 is an image using an omnidirectional lens, and the image by the omnidirectional lens may cause a blind spot in the central region.

However, as shown in the right image of FIG. 2 , the photographing unit 110 of the present invention can minimize the blind area by applying a fisheye lens, apply a lens supporting up to 5Mega (QHD), and use a 3Mega-class image sensor. It may be configured to secure a high-resolution 360-degree image by applying it.

3 is an exemplary diagram illustrating a process of processing an image in the surround IP camera 100 .

Referring to FIG. 3 , the controller 190 of the surround IP camera 100 may divide an image captured by the photographing unit 110 into a plurality of sub-regions to independently determine whether each sub-region is abnormal.

The control unit 190 determines the fire type through the color of the flame and the fire area of the image taken in the sub-region determined as having an abnormality, and predicts and alerts the fire type by pre-projecting the fire type in the sub-region located in the direction of the fire. can

In this case, the control unit 190 may calculate an estimated time for the fire to arrive from the ignition point to each sub-region after detecting the ignition point when the fire is detected. Therefore, as shown in FIG. 3 , when the fire propagation is expected to occur in the sub-region where the combustible material storage box is recognized as an object, it calculates and informs the estimated time for the fire to reach the sub-region, thereby enabling a quick response.

In particular, when detecting the ignition point, the control unit 190 may calculate the fire propagation speed for each direction. That is, as shown in Figure 3, the fire propagation speed in the right direction (V1) is the fastest, then the fire propagation speed in the upper direction (V2) is fast, and then the fire propagation speed in the lower direction (V3) is fast, Finally, it is assumed that the fire propagation speed (V4) in the left direction is detected the slowest.

In this case, the control unit 190 operates to relatively more rapidly process the image processing cycle of the sub-region in the direction in which the fire is propagating the fastest in consideration of the fire propagation speed V1 in the right direction.

That is, the controller 190 divides the image captured by the photographing unit 110 into a plurality of sub-regions and independently determines whether there is an abnormality in each sub-region, and sequentially performs object recognition and situation recognition processing of each sub-region. proceed to

Therefore, the treatment cycle of the sub-area with a slow fire propagation speed becomes longer (the treatment cycle becomes longer as the fire propagation speed slows down), and the treatment cycle of the sub-area with a high fire spread speed becomes shorter (the faster the fire propagation speed, the shorter the treatment cycle). is shortened)

In this way, object recognition of each sub-region is performed independently and the processing cycle of the sub-region selected with high priority through context recognition is shortened, thereby enabling faster and more accurate detection of changes in the situation.

That is, when the control unit 190 time-divisions a plurality of sub-regions and sequentially processes the images, basically, each sub-region is image-processed in a preset order, and when an emergency situation such as a preset situation or a fire is automatically detected, The image processing cycle of each sub-region can be individually adjusted in consideration of the speed at which danger arrives, direction, and object recognition (combustible material).

In addition, the controller 190 tracks and observes the object in the infrared image captured by the photographing unit 110 , and may determine that a problem has occurred in the object based on the body temperature distribution and body temperature change of the motionless object. Therefore, it is possible to preemptively detect the physical danger situation of the elderly, such as children and the elderly living alone.

For reference, since the control unit 190 includes an image storage memory, a communication module, and a battery module as an integrated body, even if a separate communication module is not provided, data can be transmitted/received using the built-in communication module. At this time, when a separate external communication module is provided, the control unit 190 processes data so as to transmit and receive data by selectively using a built-in communication module or an external communication module.

The vibration detection unit 120 detects a vibration pattern of the area monitored by the photographing unit 110 . That is, the controller 190 may analyze the waveform of the vibration detected by the vibration detection unit 120 to predict a risk factor corresponding to the vibration waveform. For example, it is possible to predict risk factors by detecting vibration waveforms such as a human falling sound, a human moaning sound, a shouting sound, a specific language, or an explosion sound.

The sound recognition unit 130 detects a sound in the area monitored by the photographing unit 110 . That is, the control unit 190 may analyze the sound pattern sensed through the sound recognition unit 130 to predict a risk factor corresponding to the sound pattern. For example, it is possible to predict risk factors by detecting sound patterns such as a person's moaning sound, shouting sound, specific language, or explosion sound. For reference, the control unit 190 may detect a waveform of vibration and a pattern of sound by reflecting both the detection data of the vibration detection unit 120 and the sound recognition unit 130 .

In addition, the sound recognition unit 130 is configured with a plurality of directional microphones to recognize sounds sensed in each direction. For example, when four microphones are arranged at the same location, a partition wall is installed between each microphone, which can assist in identifying the source of sound. Accordingly, when a specific voice pattern is detected from the sound collected by the sound recognition unit 130 , the control unit 190 may control the photographing unit 110 to enlarge and photograph the direction in which the sound is collected.

In addition, the control unit 190 detects the sound of gas leaking at home, the sound of falling objects, and the sound of a person falling in a specific place (bathroom), so that the photographing unit 110 enlarges the direction in which the sound is collected and shoots it. It can be controlled to output a warning sound while controlling.

4 is a block diagram of an intelligent monitoring system 1 having a surround IP camera 100 .

Referring to FIG. 4 , the intelligent monitoring system 1 includes a surround IP camera 100 , a data server 300 , and a user terminal 400 .

Since the operation of the surround IP camera 100 has been described in detail with reference to FIGS. 1 to 3 , a redundant description will be omitted.

The data server 300 receives information transmitted from the control unit 190 and provides images and detection information to at least one or more user terminals 400 registered in advance. The data server 300 automatically transmits the corresponding image and information to the pre-registered user terminal 400 whenever a pre-registered event occurs.

Here, the user terminal 400 is a generic term for devices that a user can carry and use, such as a mobile phone, a smart phone, a smart pad, and the like.

5 is an exemplary view showing an image generated by the intelligent monitoring system (1).

Referring to FIG. 5 , the controller 190 sets the area in which the object is detected in the photographing area as the gaze concentration zone (red dotted line), and when a dangerous situation of the detected object is detected, the area of the gaze concentration zone is automatically adjusted. .

For example, the control unit 190 sets and displays a gaze concentration zone with a predetermined size from the time when a preset object (baby, elderly) is detected, and when the object (baby) moves in the direction of a hot object (electric kettle) , while reducing the focus area, highlight “Baby, Hot, Dangerous Goods Phrases” in red. At this time, the information displayed regardless of the baby and dangerous situations is automatically moved out of the gaze concentration area. When the control unit 190 has a built-in display device, such information is displayed on the built-in display device, and may be displayed on the display devices of the data server 300 and the user terminal 400 .

At this time, the temperature and shape of the object (baby) and the electric port are the capturing unit 110, the vibration detecting unit 120, the sound recognition unit 130, and the LiDAR/RADAR sensor of the surround IP camera 100. It may be identified by collecting all the information of (140).

Also, the controller 190 may set an area having a predetermined temperature change in the photographing area as the gaze concentration area (red dotted line). In this case, the controller 190 may automatically adjust the area of the gaze concentration zone in proportion to the temperature change. For example, when a certain temperature abnormality is detected, a gaze concentration zone is set, and when the temperature reaches the level to cause a fire, the gaze concentration zone is concentrated in the corresponding zone, the phrase “fire caution” is additionally displayed, and the temperature is displayed in red. can be highlighted with

The data server 300 may feed back a portion of the received image in which there is no change in the image, and the controller 190 may reduce the storage capacity by automatically removing the portion in which the image does not change. Also, after dividing the image into a plurality of regions (sub regions), the data server 300 may feed back information indicating that the stored image frame for each region can be varied in consideration of the movement speed of the object.

For reference, the intelligent monitoring system 1 may capture a facial image of an object (person) recognized based on the visible image captured by the surround IP camera 100 .

The controller 190 calculates the heart rate by measuring the flow of pulsatile blood in the facial artery through the color change of the photographed facial image. That is, after selecting a region of interest among the photographed facial images, it is divided into RGB (Red Green Blue) channels, and a nonlinear scale factor is given to each of the RGB channels to adjust the signal-to-noise ratio (SNR), and the non-linear scale A heart rate is calculated based on a signal to which a scale factor is assigned.

That is, the controller 190 extracts a facial image of an object (person), selects a region of interest from among the facial images, separates them into RGB (Red Green Blue) channels, and applies a nonlinear scale factor to each of the RGB channels. to adjust a signal-to-noise ratio (SNR), and calculate a heart rate based on a signal to which a non-linear scale factor is assigned.

In addition, after converting the signal to which the non-linear scale factor is applied to the frequency domain, the controller 190 may calculate the respiration rate based on the largest power spectrum frequency value among the frequency domains corresponding to the heart rate.

That is, the controller 190 may determine the sleep state, death state, and coma possibility of the object (person) based on the calculated heart rate and respiration rate, and difficulty breathing or heart attack based on the calculated heart rate and respiration rate can even be judged.

Also, the controller 190 may warn after recognizing a correlation with each object based on the detected speed and shape of the plurality of objects. For example, when a child (object 1) and a motorcycle (object 2) are detected at the same time, the controller 190 may warn when the motorcycle moves in the direction of the child at a predetermined speed or more.

In addition, the surround IP camera 100 receives repeated reception in the same range (the same time interval between the transmission clock and the reception sampling clock) in order to suppress and receive random noise generated inside the chip, and then changes to reception in the next range. can operate to do so.

In addition, after dividing the preset photographing area into a plurality of sub-regions, the controller 190 may independently preset a detection time, a detection object, and a detection event for each area.

In addition, the data server 300 receiving the image transmitted from the controller 190 may perform an object recognition operation by itself. That is, the data server 300 feeds back the recognized object information to the control unit 190, and each object information includes an identification code allocated in advance for each object type and location information for each time of the central region of each object.

For example, when an object called an elderly person is recognized, an identification code pre-allocated to the elderly and location information for each time on the location (coordinate) of the central area of the elderly are transmitted.

For reference, the identification code includes an object code and an additional code. The object code is a code assigned to the shape of an elderly person, and the additional code is defined as encoding additional data information such as the elderly person's name.

Therefore, the controller 190 receives the object information transmitted from the data server 300 without directly processing the image processing task, and grasps the recognized name (identification of the person's name) and the movement location of the object called the elderly. can The data server 300 simultaneously recognizes a plurality of objects at the request of the controller 190 and can track their respective locations in real time.

In addition, when the intelligent monitoring system 1 detects an elderly person as an object, the elderly protection mode is executed.

In the elderly protection mode, when an elderly person is sitting on a sofa and there is a mirror reflecting the elderly person's face, the surround IP camera 100 can detect the elderly person's face only through the mirror,

The surround IP camera 100 may detect an image of a face reflected in a mirror. Therefore, the surround IP camera 100 may recognize the heart rate and respiration rate of the elderly person after object recognition of the face reflected in the mirror.

For reference, when an elderly person is recognized as an object and the elderly protection mode is executed, an additionally provided motion sensor periodically transmits and receives a radio wave of a predetermined wavelength in the direction in which the object is recognized, so that the movement of the object can be more accurately identified. .

That is, when an elderly person collapses on the back of the sofa, he may be out of the image by the sofa, so a motion sensor is additionally used to compensate for this.

The surround IP camera according to an embodiment of the present invention tracks and observes an object in an infrared image, but can automatically determine that a problem has occurred in the object based on the body temperature distribution and body temperature change of the non-moving object.

In addition, the surround IP camera can divide the captured image into a plurality of sub-regions and automatically determine whether there is an abnormality in each sub-region independently.

As such, those skilled in the art to which the present invention pertains will understand that the present invention may be embodied in other specific forms without changing the technical spirit or essential characteristics thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100: Surround IP camera
110: filming unit
120: vibration detection unit
130: sound recognition unit
140: lidar (LiDAR) / radar (RADAR) sensor
190: control unit
300 : data server
400: user terminal

Claims (6)

A photographing unit capable of omnidirectional imaging and omnidirectional infrared imaging;
a control unit for tracking and observing a subject in the infrared image captured by the photographing unit, and determining that a problem has occurred in the subject based on the body temperature distribution and body temperature change amount of the non-moving subject;
a communication unit for transmitting a result determined through the control unit; and
a vibration detection unit for detecting vibration; and
The control unit divides the image captured by the photographing unit into a plurality of sub-regions and independently determines whether each sub-region has an abnormality, to determine the type of fire through the
The control unit predicts a risk factor corresponding to the vibration waveform by analyzing the waveform of the vibration detected through the vibration detection unit,
The control unit calculates an estimated time for a fire to reach each sub-region from the fire point after detecting an ignition point upon detection of a fire, but controls the image processing cycle of the sub-region having a relatively slow fire propagation speed to be longer, and fire propagation The image processing cycle of the relatively fast sub-region is controlled so that it becomes faster,
The data server receives the image transmitted from the control unit and performs an object recognition operation by itself, but feeds back object information including identification codes allocated in advance for each object type and location information for each object's central region by time to the control unit Surround IP camera, characterized in that. delete delete delete delete delete

Images