KR102788241B1 - Nursing Facility Control System and Method for Protecting Privacy - Google Patents

Abstract

A nursing facility control system and method capable of protecting privacy are disclosed.
According to one aspect of the present embodiment, a control system is provided, characterized by including an image surveillance device that captures a surrounding area, detects a subject in the image, and masks the face area of the subject, and an image processing device that unmasks a specific person in the image captured by the image surveillance device and provides the unmasked person to the outside.

Description

Nursing Facility Control System and Method for Protecting Privacy

This embodiment relates to a nursing facility control system and method capable of protecting the privacy or personal information of a patient.

The material described in this section merely provides background information for the present embodiment and does not constitute prior art.

Most of the patients admitted to nursing homes are elderly or seriously ill and have difficulty moving on their own, so they need a lot of help from nurses and caregivers. However, the number of nurses and caregivers in nursing homes is significantly smaller than the number of patients. Therefore, nurses and caregivers need to care for and check the patients' conditions 24 hours a day, but in reality, it is difficult to fully perform this.

It is not realistically possible to assign nursing caregivers to each patient due to the excessive demand for manpower and high labor intensity. To compensate for this, plans are being discussed to deploy CCTVs and conduct constant monitoring from a central control room.

However, since it is not enough to check all situations with a small number of personnel, the reality is that when an accident actually occurs, the response is delayed and a great danger is encountered. Therefore, if there is no caregiver or nurse nearby in an emergency situation such as a patient falling out of bed or collapsing while moving, there is a constant problem that immediate response is not possible and there is a possibility that a major accident may occur.

Recently, there have been some unfortunate cases in which elderly or seriously ill patients have been unable to resist or explain the damage to their guardians due to the harsh treatment of nursing staff, and conversely, there have been various problems occurring in nursing homes, such as disputes between nursing homes and guardians arising due to injuries caused by the patient's negligence or false statements due to illusions or mental disorders, even when there has been no harsh treatment.

In addition, CCTVs are deployed to solve the aforementioned problems, but personal information such as the faces of nurses, caregivers, and patients are directly exposed in the footage captured by CCTV, which poses a vulnerability in terms of personal information protection.

One embodiment of the present invention aims to provide a nursing facility control system and method that protects privacy by masking the face or personal information of a person appearing in a video in real time.

According to one aspect of the present embodiment, a control system is provided, characterized by including an image surveillance device that captures a surrounding area, detects a subject in the image, and masks the face area of the subject, and an image processing device that unmasks a specific person in the image captured by the image surveillance device and provides the unmasked person to the outside.

According to one aspect of the present embodiment, the object is characterized in that it is an object or a face of an object.

According to one aspect of the present embodiment, the target detection unit is characterized in that it uses both an algorithm for detecting the object and an algorithm for detecting the face of the object, and detects the one that is detected with priority among the two as the target.

According to one aspect of the present embodiment, the video surveillance device is characterized in that it detects an object or a face of the object by using either an algorithm for detecting the object or an algorithm for detecting the face of the object.

According to one aspect of the present embodiment, the masking unit is characterized by masking a preset area of an end in the direction of the head of the detected object when the object is detected.

According to one aspect of the present embodiment, the masking unit is characterized by covering the face of the object with a masking box having a preset area when the face of the object is detected.

According to one aspect of the present embodiment, the control system is characterized by further including some or all of a distance sensor for sensing a distance between the control system and a preset point within a shooting location and a pressure sensor for sensing a pressure applied to the control system.

According to one aspect of the present embodiment, the distance sensor is characterized in that it is implemented with some or all of a lidar sensor, a radar sensor, and a depth sensor.

According to one aspect of the present embodiment, the image processing device is characterized by receiving an image or metadata of a specific person for unmasking from the outside.

According to one aspect of the present embodiment, the image processing device is characterized by unmasking a specific input person.

According to one aspect of the present embodiment, the video surveillance device is characterized by including a photographing unit, a target detection unit that detects an object or a face of an object in an image photographed by the photographing unit, and a masking unit that masks a face area of an object detected by the target detection unit.

As described above, according to one aspect of the present embodiment, there is an advantage in that privacy can be protected by masking the face or personal information of a person appearing in a video in real time.

In addition, according to one aspect of the present embodiment, there is an advantage in that an abnormality can be detected in a patient or a space where a patient resides.

FIG. 1 is a drawing illustrating a nursing facility control system according to one embodiment of the present invention.
FIG. 2 is a drawing illustrating the configuration of a video surveillance device according to the first embodiment of the present invention.
FIG. 3 is a diagram illustrating the configuration of an image processing device according to one embodiment of the present invention.
FIG. 4 is a flowchart illustrating a method for real-time masking by a video surveillance device according to one embodiment of the present invention.
FIG. 5 is a flowchart illustrating a method for an image processing device to recognize a face in an image according to one embodiment of the present invention.
FIG. 6 is a flowchart illustrating a method for an image processing device according to one embodiment of the present invention to remove masking performed on a specific person in an image.
FIG. 7 is a diagram illustrating the configuration of a video surveillance device according to the second embodiment of the present invention.

The present invention can have various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, but should be understood to include all modifications, equivalents, or substitutes included in the spirit and technical scope of the present invention. In describing each drawing, similar reference numerals are used for similar components.

The terms first, second, A, B, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are only used to distinguish one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term and/or includes any combination of a plurality of related described items or any item among a plurality of related described items.

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

The terminology used in this application is only used to describe specific embodiments and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. It should be understood that the terms "comprise" or "have" in this application do not exclude in advance the possibility of the presence or addition of features, numbers, steps, operations, components, parts or combinations thereof described in the specification.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and will not be interpreted in an idealized or overly formal sense, unless expressly defined in this application.

In addition, each configuration, process, procedure or method included in each embodiment of the present invention may be shared within a scope that is not technically contradictory to each other.

FIG. 1 is a drawing illustrating a nursing facility control system according to one embodiment of the present invention.

Referring to FIG. 1, a nursing facility control system (100) according to one embodiment of the present invention includes a video surveillance device (110), a video storage device (120), an video processing device (130), and a control center (140). In addition, the nursing facility control system (100) may further include a sensor (not shown).

The nursing facility control system (100) is a system that films the places where patients, nursing caregivers, and other related persons stay or work (hereinafter, abbreviated as “activity places”) and masks the faces of patients or related persons in real time in the video to protect the privacy of patients or related persons.

A video surveillance device (110) is installed at an activity site to film within a certain radius, and at the same time, recognizes an object or an object's face in the filmed image and masks the face or the recognized facial part within the object. A representative example of a video surveillance device (110) is a CCTV, and is installed for the purpose of detecting an abnormality in an object, an intrusion of an outsider into a specific space, or securing evidence in various situations, and continuously films its surroundings. At this time, the video surveillance device (110) recognizes an object or a face (of an object) in real time within the image being filmed, and masks the face or the recognized facial part within the object in real time so that post-processing may not be performed in a separate device.

The video surveillance device (110) converts the face (of the object) recognized in the video into metadata and transmits the original video, masked video, and metadata to the video storage device (120). A more specific description of the video surveillance device (110) will be described later with reference to FIG. 2.

The video storage device (120) stores the original video, masked video, and metadata received from the video surveillance device (110). The video storage device (120) performs a local area connection with the video surveillance device (110), and receives and stores the original video captured by the video surveillance device (110), the masked video in which the face area of the object is masked in real time, and metadata about the recognized face. The video storage device (120) can be located within a preset radius from the video surveillance device (110), and can include a security element for storing the video and metadata. The video storage device (120) can be implemented as an NVR (Network Video Recorder), etc.

The image processing device (130) communicates with the image storage device (120), views or manipulates the image stored in the image storage device (120), and releases the masking of a specific person and provides it to the outside.

The image processing device (130) can also be connected to the image surveillance device (110) and the image storage device (120) within a local area. Alternatively, the image processing device (130) can be connected to the image storage device (120) in an external space through a VPN (Virtual Private Network). Accordingly, the image processing device (130) can access the image stored in the image storage device (120) while maintaining security. Accordingly, the image processing device (130) can view the image stored in the image storage device (120) or manipulate the image.

The image processing device (130) can receive an external request to unmask a specific person, unmask the person, and provide the corresponding image to the outside. For the purpose of securing evidence against a criminal, an image of a specific person can be requested from the outside to the image processing device (130). Accordingly, the image processing device (130) can receive an image or metadata of a specific person from the outside, and analyze metadata of the specific person's face from the received image of the specific person. The image processing device (130) searches for metadata of the specific person in the image and determines whether the specific person appears in the image. If the specific person appears in the image, the image processing device (130) unmasks the person and provides the image to the outside. The specific configuration of the image processing device (130) will be described later with reference to FIG. 3.

The control center (140) receives a notification from the video surveillance device (110) and analyzes whether there is an abnormality in the patients in the activity site. The control center (140) receives the captured images from the video processing device (130) and outputs them. Accordingly, the relevant personnel can use the control center (140) to determine whether the patients in each activity site are doing well.

Although Fig. 1 shows that the nursing facility control system (100) includes a video storage device (120), it is not necessarily limited thereto. Without the video storage device (120), the video surveillance device (110) and the video processing device (130) can be connected via VPN and directly transmit and receive video and data.

FIG. 2 is a drawing illustrating the configuration of a video surveillance device according to the first embodiment of the present invention.

Referring to FIG. 2, the video surveillance device (110) according to the first embodiment of the present invention includes a photographing unit (210), a target detection unit (215), a masking unit (220), a communication unit (225), an image color conversion unit (230), a brightness value equalization unit (235), a pixel value change calculation unit (240), a face image recognition unit (245), and a control unit (250). In addition, the video surveillance device (110) may further include a monitoring unit (255).

The camera unit (210) continuously photographs the area within the range it can photograph from the installed location. The camera unit (210) can continuously photograph a certain range in a fixed state, or can rotate and continuously photograph a certain range.

The target detection unit (215) detects a detection target in the image captured by the photographing unit (210). The detection target may be an object or a face of an object. The target detection unit (215) may use both an algorithm for detecting an object and an algorithm for detecting a face of an object for rapid target detection. Alternatively, the target detection unit (215) may use one algorithm selected by the masking unit (220). That is, the target detection unit (215) may operate both algorithms and detect a detected object with priority, or may use one algorithm to detect only an object or a face of an object.

The target detection unit (215) can detect an object through the following process. The target detection unit (215) detects an object based on changes in pixel values within an image. In the case of a person, movement is inevitable. Movement is inevitable not only when moving, but also when sitting or standing still, even when a body part such as an eyelid is moving. The target detection unit (215) primarily detects an object based on whether there is a change in pixel values for a specific object in an image (between a frame at a specific time and a frame after that time).

However, there is no guarantee that all objects are people just because there is a change in pixel values in the image. For example, there may be cases where plants move due to the wind, or they may be animals. Accordingly, the object detection unit (215) additionally checks the distribution of pixel values of the detected object. Typically, the pixel distribution of humans has a certain tendency regardless of race, age, and gender, and this is clearly different from plants and animals. Accordingly, the object detection unit (215) detects objects based on whether there is even a small change in pixel values and whether the distribution of pixel values has a human shape.

At this time, the target detection unit (215) can detect whether the object in the image is a person by using a learning model learned with a preset learning model, with the pixel value of the person as an input value and whether the object in the image is a person as an output value. As an example of a representative learning model, CNN (Convolutional Neural Networks) can be used. CNN performs learning on various images based on the input and output values described above. The target detection unit (215) can infer (Inference) whether the detected object is a person based on the distribution of pixel values using the learned learning model.

The target detection unit (215) can directly detect the face of an object through the following process. With the input value as an image or video and the output value as the face of an object in the video, numerous learning data can be trained using a learning model such as CNN. The target detection unit (215) can directly detect the face of an object in the video, including an artificial intelligence model or machine learning model that has been trained in this way.

Additionally, the target detection unit (215) can analyze the movement of the target by analyzing the change in the pixels of the target after detecting the target. The target detection unit (215) can analyze the change in the pixel value of the detected target as a vector. The target detection unit (215) can analyze the speed of the target based on how much a specific pixel in the detected target has moved per unit time. The target detection unit (215) can transmit the speed of the analyzed target to the masking unit (220).

The masking unit (220) performs masking on an object detected from the detection result of the target detection unit (215) or on the face of the detected object. In order to secure speed, the masking unit (220) may perform masking on an object detected by the target detection unit (215) with priority. Alternatively, the masking unit (220) may allow the target detection unit (215) to selectively detect either the object or the face of the object in order to improve convenience, and may perform masking on an object detected by the target detection unit (215).

The masking unit (220) performs masking on the object detected by the target detection unit (215). At this time, if the object is detected first, the masking unit (220) recognizes a face within the object in order to perform rapid masking, and performs masking not only on the face, but also masks an area equal to a preset area at the end in the direction of the head of the detected object. Here, the preset area is determined as the product of a length equal to a preset ratio of the total length of the detected object and a width equal to or greater than the width of the detected object. For example, considering the ratio between the height and head size of a human, the masking unit (220) forms a masking area having a length corresponding to 1/10 to 1/6 of the total length of the detected object and a width equal to or greater than the width of the detected object, and places it at the end in the direction of the head of the detected object. Accordingly, the masking unit (220) can perform masking on the faces of all objects while ensuring real-time performance. Masking is performed on the head (face) area of all objects in the video.

At this time, if masking is performed indiscriminately according to the ratio of the object, the following problem may occur. For example, a sitting adult and a standing child may have similar overall (height-wise) lengths. However, since the sizes of the heads of an adult and a child are different, if a masking area is formed indiscriminately according to the ratio, masking may not be performed completely on the head of a sitting adult. Accordingly, the masking unit (220) determines whether the height-wise length of the detected object exceeds a preset reference value. Here, the preset reference value may be the average sitting height of an adult male. If the height-wise length of the detected object exceeds the preset reference value, the masking unit (220) masks the object by a preset ratio of the overall length. Conversely, if the height-wise length of the detected object does not exceed the preset reference value, the masking unit (220) masks the object by a ratio greater than the preset ratio. For example, if the preset ratio is 1/10 to 1/6 of the total length of the detected object, the masking unit (220) performs masking for 1/4 to 1/2 of the total length of the object in the aforementioned situation. Accordingly, the masking unit (220) can completely mask the head (face) of the object in any situation or posture.

Meanwhile, the masking unit (220) masks the face of the detected object. The masking unit (220) does not mask an area equal to the face of the detected object, but masks the face of the object by covering it with a masking box having a preset area. The preset area means an area that is larger than a preset ratio of the size of a face that a person can typically have. In this way, the masking unit (220) masks the face of the object by covering the face of the detected object with a masking box having a preset area in order to secure real-time performance.

The masking unit (220) can continuously mask the target by predicting the target's movement path based on the target's speed information received from the target detection unit (215). The target's movement path can be predicted based on the target's speed. Accordingly, the masking unit (220) can continuously mask each target within the image by moving the masking area together with the object's movement.

The communication unit (225) transmits the captured image of the shooting unit (210), the masking image of the masking unit (220), and the metadata of the face recognized by the face image recognition unit (245) described later to the image storage device (120).

The image color conversion unit (230) converts the colors of the original image into black and white. In image processing, color images can take up excessive capacity. To solve this problem, the image color conversion unit (230) converts the colors of the original image into black and white.

The brightness equalization unit (235) equalizes the brightness values (brightness values) of the image converted to black and white. In some cases, the image itself may be relatively dark (in terms of brightness) or relatively bright. In order to make the results more accurate in the process of comparing face IDs within the image in the future, the brightness equalization unit equalizes the brightness values of the image converted to black and white. Accordingly, the brightness values of all images can be made constant.

The pixel value change calculation unit (240) calculates the change in pixel values within the image. The change in pixel values within the image varies depending on the shape of the facial contour or facial features of the object (person). The pixel value change calculation unit (240) calculates the change in pixel values within the image where the brightness values are equalized.

At this time, the pixel value change calculation unit (240) can decompose the image into blocks and calculate the pixel value change for each block. The pixel value change calculation unit (240) analyzes the pixel value change for each block, especially for adjacent blocks.

The facial image recognition unit (245) recognizes each individual's face based on the change in pixel value for each block within the image and assigns a face ID to each of them.

The face image recognition unit (245) can recognize the position of a face in an object (person) based on the change in pixel values for each block in the image. The face image recognition unit (245) recognizes faces in each object in the image and assigns different IDs to each face. The face image recognition unit (245) assigns different IDs to different faces of each individual, thereby enabling the face of a specific person to be identified and managed by ID.

At this time, the face image recognition unit (245) may convert the face image into metadata to facilitate management and search in the image. The face image recognition unit (245) uses various parameters in the image of the object, such as changes in pixel values for each block in the image, to generate metadata that allows for recognizing and detecting a specific face ID for each face image.

As described above, the video surveillance device (110) recognizes a face image in an image converted to black and white anyway. Accordingly, the video surveillance device (110) can detect an object or an object's face more smoothly in a situation where sunlight or light is shining, but can detect an object or an object's face without difficulty even in a situation where sunlight or light is not shining, such as at night or in the early morning.

The control unit (250) controls the aforementioned operations of each configuration.

The control unit (250) controls the communication unit (225) to transmit the original image, the masked image, and the metadata to the image storage device (120), and may encrypt and transmit the original image, the masked image, and the metadata. The image storage device (120) may store the encrypted data in an encrypted state, and may perform decryption when the image processing device (130) accesses the image for viewing or manipulating the image.

The monitoring unit (255) outputs the original image or the masked image under the control of the control unit (250). The monitoring unit (255) allows the relevant persons to view the image captured by the shooting unit (210) or the image masked by the masking unit (220) on site.

FIG. 3 is a diagram illustrating the configuration of an image processing device according to one embodiment of the present invention.

Referring to FIG. 3, an image processing device (130) according to one embodiment of the present invention includes a communication unit (310), a memory unit (320), an input unit (330), a masking release unit (340), and a control unit (350). In addition, the image processing device (130) may further include a monitoring unit (360).

The communication unit (310) communicates with the video surveillance device (110) and receives information on whether an abnormality has occurred or whether an outsider has intruded from the video surveillance device (110). Under the control of the control unit (350), the communication unit (310) transmits information on whether an abnormality has occurred or whether an outsider has intruded at the activity site to the control center (140).

The communication unit (310) transmits the image unmasked by the masking unit (340) to the outside.

The communication unit (310) communicates with the video storage device (120) via VPN and accesses or receives videos stored in the video storage device (120). In addition, the communication unit (310) receives metadata stored in the video storage device (120).

The memory unit (320) stores images and metadata received from the image storage device (120) by the communication unit (310).

The input unit (330) receives an image or metadata of a specific person for unmasking from the outside. The input unit (330) mainly receives metadata of a specific person for unmasking, and transmits the received metadata to the unmasking unit (340).

The unmasking unit (340) unmasks a specific person input from the input unit (303). The unmasking unit (340) compares the metadata of the specific person received from the outside with the metadata of each person appearing in the video, stored in the memory unit (320). In order to confirm whether the specific person appears in the video, the unmasking unit (340) compares the metadata of the specific person with the metadata of each object (person) stored in the memory unit (320). The fact that the metadata of the specific person is stored in the memory unit (320) means that the specific person appears in the video. Accordingly, the unmasking unit (340) determines whether the specific person appears in the video based on the comparison result. The unmasking unit (340) unmasks the specific person in the video based on the comparison result. Unmasking a specific person in the video of the unmasking unit (340) can be carried out by unmasking only the specific person, or by masking all people except the specific person in the original video.

The control unit (350) controls the operation of each component in the image processing device (130).

The monitoring unit (360) can output an original image or a masked image under the control of the control unit (350).

FIG. 4 is a flowchart illustrating a method for real-time masking by a video surveillance device according to one embodiment of the present invention.

The target detection unit (215) detects a target in an image captured by the shooting unit (210) (S410).

The masking unit (220) first determines whether the detected object is an object (S420).

First, if the detected target is an object, the masking unit (220) masks a preset area from one end of the object (S430).

First, if the detected object is a face of an object, the masking unit (220) masks the detected face with a masking box having a preset area (S440).

The target detection unit (215) determines whether the detected target is moving (S450).

The target detection unit (215) determines the speed of the target by judging pixel changes (S460).

The masking unit (220) moves the masking area based on the determined speed (S470). At this time, the masking unit (220) can intentionally increase the area of the masking area.

The communication unit (225) sends the masked image and the original image to the image storage device (120) (S480).

However, without determining whether the detected target is an object (S420), the masking unit (220) can select either the object or the face of the object and immediately perform masking (S430 process or S440 process) on the selected target.

FIG. 5 is a flowchart illustrating a method for an image processing device to recognize a face in an image according to one embodiment of the present invention.

The target detection unit (215) detects a target in the original image (S510).

The image color conversion unit (230) converts the color of the image to black and white (S520).

The brightness equalization unit (235) equalizes the brightness values of the converted image (S530).

The pixel value change calculation unit (240) decomposes the equalized image into blocks and calculates the pixel value change for each block (S540).

The facial image recognition unit (245) recognizes each individual's face based on the generated value and converts it into metadata (S550).

FIG. 6 is a flowchart illustrating a method for an image processing device according to one embodiment of the present invention to remove masking performed on a specific person in an image.

The input unit (330) receives metadata of a person for unmasking (S610).

The masking release unit (340) determines whether metadata of the input person exists in the stored metadata (S620).

If metadata of the input person exists in the saved metadata, the masking unmasking unit (340) searches for metadata of the corresponding person in the saved masked image (S630).

The masking release unit (340) releases the mask of the person in question (S640).

The communication unit (310) sends a masked video of the person in question with the masking removed (S650).

FIG. 7 is a diagram illustrating the configuration of a video surveillance device according to the second embodiment of the present invention.

Referring to FIG. 7, a video surveillance device (110) according to a second embodiment of the present invention includes an object detection unit (710) and a face detection unit (720) instead of the target detection unit (215) in the video surveillance device (110) according to the first embodiment of the present invention. The photographing unit (210), the communication unit (225), the image color conversion unit (230), the brightness value equalization unit (235), the pixel value change calculation unit (240), the face image recognition unit (245), the control unit (250), and the monitoring unit (255) in the video surveillance device (110) according to the second embodiment of the present invention perform the same operations as those in the video surveillance device (110) according to the first embodiment of the present invention.

The object detection unit (710) performs the same object detection operation as the target detection unit (215) in the image captured by the shooting unit (210).

The face detection unit (720) performs the same face detection operation of the object detection unit (215) in the image captured by the camera unit (210).

The object detection unit (710) and the face detection unit (720) are implemented with different configurations and perform the above-described operations in parallel.

The masking unit (220) performs masking in the same manner as that in the video surveillance device (110) according to the first embodiment of the present invention. At this time, the masking unit (220) may perform masking by receiving a detection result from the detection unit that has first detected the target it wants to detect among the object detection unit (710) and the face detection unit (720). Alternatively, the masking unit (220) may select either of the two detection units (710, 720) and receive a detection result from the corresponding detection unit to perform masking.

Although each process is described as being executed sequentially in FIGS. 4 to 6, this is merely an illustrative description of the technical idea of one embodiment of the present invention. In other words, a person having ordinary skill in the art to which one embodiment of the present invention belongs may change the order described in each drawing without departing from the essential characteristics of one embodiment of the present invention, or may modify and modify and apply various modifications and variations such as executing one or more of each process in parallel. Therefore, FIGS. 4 to 6 are not limited to a chronological order.

Meanwhile, the processes illustrated in FIGS. 4 to 6 can be implemented as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices that store data that can be read by a computer system. That is, the computer-readable recording medium includes storage media such as magnetic storage media (e.g., ROM, floppy disk, hard disk, etc.) and optical reading media (e.g., CD-ROM, DVD, etc.). In addition, the computer-readable recording medium can be distributed to computer systems connected to a network, so that the computer-readable codes can be stored and executed in a distributed manner.

The above description is merely an illustrative description of the technical idea of the present embodiment, and those with ordinary skill in the art to which the present embodiment belongs may make various modifications and variations without departing from the essential characteristics of the present embodiment. Therefore, the present embodiments are not intended to limit the technical idea of the present embodiment, but to explain it, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The protection scope of the present embodiment should be interpreted by the following claims, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of the rights of the present embodiment.

100: Nursing home control system
110: Video surveillance device
120: Video storage device
130: Image processing device
140: Control Center
150: Bed
210: Filming Department
215: Target detection unit
220: Masking Department
225, 310: Communications Department
230: Image color conversion section
235: Lightness equalization section
240: Pixel value change calculation section
245: Facial image recognition unit
250, 350: Control Unit
255, 360: Monitoring Department
320: Memory section
330: Input section
340: Unmasking section
710: Object detection unit
720: Face detection unit

Claims (9)

A video surveillance device that captures the surroundings, detects an object or the face of an object in the video, and masks the face area of the object; and
Includes an image processing device that removes the masking of a specific person in the image captured by the above image surveillance device and provides it to the outside.
The above video surveillance device,
filming department;
A target detection unit that detects an object or the face of an object by using both an algorithm for detecting an object and an algorithm for detecting a face of the object, and preferentially detects an object or the face of the object in an image captured by the camera unit, or detects an object or the face of the object by using only one of the algorithm for detecting an object and the algorithm for detecting a face of the object; and
It includes a masking unit that masks an object or the face of an object detected by the above target detection unit,
The above masking unit masks a preset first area of the end of the head direction of the detected object when the object is preferentially detected by the target detection unit.
The above masking unit masks the face of the object by covering it with a masking box having a preset second area when the face of the object is preferentially detected by the target detection unit.
A control system, characterized in that the above-described first area is determined as the product of a length equal to a preset ratio of the total length of the detected object and a numerical value greater than the width of the detected object. In the first paragraph,
The above image processing device,
A control system characterized by receiving an image or metadata of a specific person for unmasking from the outside. In the second paragraph,
The above image processing device,
A control system characterized by unmasking a specific input person.



delete delete delete delete delete delete

Images