KR102336480B1 - Image Processing Apparatus for Automatic Segment Classification and Driving Method Thereof - Google Patents

Abstract

The present invention relates to an image processing apparatus for automatically classifying segments and a driving method thereof. According to an embodiment of the present invention, the image processing apparatus for automatically classifying segments may comprise: a communication interface unit receiving a photographed image of an arbitrary space; and a control unit segmenting and automatically classifying a plurality of regions wherein designated different types of objects including a place in the received photographed image are included into each region and processing an event for each segment based on the automatically classified segment.

Description

Image Processing Apparatus for Automatic Segment Classification and Driving Method Thereof

The present invention relates to an image processing apparatus for automatic segment classification and a driving method of the apparatus, and more particularly, it is possible to determine an event related to each segment through segmentation of a specified object included in a background in a captured image, for example. It relates to an image processing apparatus for automatic segment classification and a driving method of the apparatus.

Recently, a large number of surveillance cameras have been installed and operated simultaneously in public places, offices, houses, and crime-prone areas for the purpose of preventing crime and enhancing security. Each image information obtained by being photographed in real time through such a large number of surveillance cameras is displayed on a control monitor provided inside the control room. At this time, the screen of the control monitor is divided into hardware or software to correspond to the number of surveillance cameras, and each image information is displayed on the control monitor to be distinguished from each other.

In this regard, in the prior art, since it is impossible for a limited number of control personnel to control all CCTV camera images, intelligent control is performed so that the control personnel can check only the necessary, that is, images of a specific event, or the area of interest (ROI) in the monitoring area. ) and monitoring only the events in the area of interest, the technology is improving in a way that reduces the fatigue caused by the control of the control personnel.

However, in the conventional ROI method, when there are many types of events, repetitive setting tasks are increased, so setting is cumbersome, and when an event occurs in an area other than the designated ROI, there is a problem in that it cannot be detected. In addition, there is a problem in that it is difficult to recognize complex situations occurring in areas other than the ROI because the purpose is to confirm the event result.

Korean Patent Publication No. 10-2018-0089977 (2018.08.10) Korean Patent Publication No. 10-2020-0061747 (2020.06.03) Korea Patent Publication No. 10-1856733 (2018.05.03) Korea Patent Publication No. 10-2028147 (2019.09.26) Korean Patent Publication No. 10-2154610 (2020.09.04) Korea Patent Publication No. 10-1668303 (2016.10.17)

An embodiment of the present invention provides an image processing apparatus for automatic segment classification capable of determining an event related to each segment through segmentation of a specified object included in a background in a captured image, and a method of driving the apparatus, for example. There is a purpose.

An image processing apparatus for automatic segment classification according to an embodiment of the present invention includes a communication interface unit for receiving a captured image of an arbitrary space, and a plurality of objects including designated different types of objects including a place in the received captured image. and a control unit for automatically classifying the segmented area into each area, and processing an event for each segment based on the automatically classified segment.

The controller may determine an event for a designated object in the arbitrary space based on the captured image, segment a background image excluding the designated object into each region, and re-determine the determined event for each segment .

The control unit may perform the segmentation by detecting at least one of a housing complex, a factory, a livestock farm, a construction site, a shopping mall, an office, a road, and a station as the different types of objects.

The controller may re-determine a complex situation based on the automatically classified segment for the event determined in relation to the designated object.

The control unit may present a selectable event type for each automatically classified segment to the user, receive an event that the user wants to detect and an event occurrence condition related to the segment, respectively, and perform an event determination operation for each segment .

The controller may receive information about the type, size, and time of an object as the event occurrence condition, respectively.

The controller may adjust the priority of events occurring for each segment by using metadata and big data generated in relation to an arbitrary object in the captured image.

In addition, the method of driving an image processing apparatus for automatic segment classification according to an embodiment of the present invention comprises the steps of: a communication interface unit receiving a captured image of an arbitrary space; and automatically classifying a plurality of regions including different types of objects by segmenting them into regions, and processing an event for each segment based on the automatically classified segment.

The processing includes determining an event for a designated object in the arbitrary space based on the captured image, segmenting a background image excluding the designated object into each region, and re-judging the determined event for each segment can

In the processing, the segmentation may be performed by detecting at least one of a housing complex, a factory, a livestock site, a construction site, a shopping mall, an office, a road, and a station as the different types of objects.

The processing may include re-judging a complex situation based on the automatically classified segment for the event determined in relation to the designated object.

The processing may include presenting a selectable event type for each automatically classified segment to the user, receiving an event and event occurrence condition that the user wants to detect in relation to the segment, respectively, and performing an event determination operation for each segment. can

In the processing, the type, size, and time information of the object may be respectively set as the event occurrence condition.

In the processing, the priority of the event occurring for each segment may be adjusted using metadata and big data generated in relation to an arbitrary object in the captured image.

According to an embodiment of the present invention, an event may be presented based on an event suitable for a segment that is an object of interest or a space, and the event may be determined.

In addition, in the embodiment of the present invention, a simple setting operation can be made even if there are many types of events, and since the entire area is analyzed without setting an ROI as in the prior art, it will be possible to recognize a complex situation in which an event occurs.

Furthermore, the embodiment of the present invention may improve false positives and false positives when determining an event by converting metadata related to a tracking object into big data.

1 is a view showing an automatic segment classification system according to an embodiment of the present invention;
2 is a block diagram illustrating a detailed structure of the image processing apparatus of FIG. 1;
3 is a block diagram illustrating a detailed structure of the segment processing unit of FIG. 2;
4 is a flowchart illustrating a driving process of an image processing apparatus according to an embodiment of the present invention;
5 is a flowchart illustrating a driving process of an image processing apparatus according to another embodiment of the present invention.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

1 is a diagram illustrating an automatic segment classification system according to an embodiment of the present invention.

As shown in Fig. 1, the automatic segment classification system 90 according to the embodiment of the present invention is a part or includes all

Herein, "including some or all" means that the communication network 110 of FIG. 1 is omitted so that the photographing apparatus 100 and the image processing apparatus 120 perform direct communication (eg, P2P communication), or the image processing apparatus Part or all of the components constituting the communication network 110 may be integrated into a network device (eg, a wireless switching device, etc.) constituting the communication network 110 and may be configured to help a sufficient understanding of the invention It will be described as including everything for the purpose.

The photographing apparatus 100 includes a Closed Circuit Television (CCTV) or Internet Protocol (IP) camera installed at any place requiring monitoring. In addition, the photographing apparatus 100 may include a PTZ (Pan-Tilt-Zoom) camera capable of pan, tilt, and zoom operations as well as a fixed camera. The photographing device 100 is installed in various places for social safety, crime prevention, social issue, and public surveillance and converts the photographed image to the image processing device 120 . can provide For example, the photographing device 100 may be installed in a public place such as a subway or bus stop to monitor various situations such as incidents and accidents, and may also monitor acts of violence occurring on a railing of a bridge or in a remote place. Furthermore, monitoring can be made through CCTV installed in daycare centers and the like. When the image processing device 120 operates as a control device, the photographing device 100 may operate under the control of the control device.

The communication network 110 includes both wired and wireless communication networks. For example, as the communication network 110 , a wired or wireless Internet network may be used or interlocked. Here, the wired network includes an Internet network such as a cable network or a public telephone network (PSTN), and the wireless communication network includes CDMA, WCDMA, GSM, Evolved Packet Core (EPC), Long Term Evolution (LTE), Wibro network, etc. is meant to include Of course, the communication network 110 according to the embodiment of the present invention is not limited thereto, and may be used, for example, in a cloud computing network under a cloud computing environment, a 5G network, and the like. For example, when the communication network 110 is a wired communication network, the access point in the communication network 110 can connect to a switching center of a telephone company, etc., but in the case of a wireless communication network, it accesses the SGSN or GGSN (Gateway GPRS Support Node) operated by the communication company to transmit data. or by accessing various repeaters such as a Base Transmissive Station (BTS), NodeB, and e-NodeB to process data.

The communication network 110 may include an access point (AP). Here, the access point includes a small base station, such as a femto or pico base station, which is often installed in a building. Femto or pico base stations are classified according to the maximum number of access to the imaging device 100, etc. in the classification of small base stations. Of course, the access point may include a short-distance communication module for performing short-distance communication such as Zigbee and Wi-Fi with the photographing device 100 and the like. The access point may use TCP/IP or Real-Time Streaming Protocol (RTSP) for wireless communication. Here, short-distance communication may be performed in various standards such as Bluetooth, Zigbee, infrared, radio frequency (RF) such as ultra high frequency (UHF) and very high frequency (VHF), and ultra-wideband communication (UWB) in addition to Wi-Fi. Accordingly, the access point may extract the location of the data packet, designate the best communication path for the extracted location, and transmit the data packet to the next device, for example, the image processing apparatus 120 along the designated communication path. The access point may share several lines in a general network environment, and includes, for example, a router, a repeater, and a repeater.

The image processing apparatus 120 may be, for example, an image analysis apparatus as a server, and may perform an operation for automatic segment classification according to an embodiment of the present invention. In this regard, the image processing apparatus 120 may be variously named, such as an automatic segment classification apparatus or an image apparatus for automatic segment classification. Of course, the image processing apparatus 120 according to the embodiment of the present invention may perform an image analysis operation in addition to the automatic image segment classification operation or operate as a control apparatus so that a controller can control the image. The image processing apparatus 120 according to an embodiment of the present invention may analyze a photographed image received from the photographing apparatus 100 to detect and classify a designated object, track the object, and determine an event based on the tracking result. . Here, the designated object may be a person object or an object of interest in which the user is interested. The image processing apparatus 120 may generate metadata for these designated objects and events and manage them in the DB 120a of FIG. 1 . Here, it can be seen that the metadata includes attribute information (eg, shape, color, etc.) related to the designated object.

Above all, the image processing apparatus 120 according to an embodiment of the present invention does not simply determine an event based on the object tracking result as in the prior art, but rather for the entire area of the shooting projection or for objects other than a designated object. That is, after setting a plurality of areas in the background of the captured image, segments can be classified into segments, or fragments, into each area, and events are set for each classified segment. A comprehensive judgment of the event is made. Alternatively, an event set by the user for each segment may be detected. For example, the image processing apparatus 120 performs segmentation on a specific object in the foreground image by performing foreground processing after excluding the specified object from the captured image. And after classifying the segments, you can set events for each segment. For example, segment classification of a background region for setting a non-ROI may be performed.

In the automatic segment classification of the background in the captured image, for example, segments for housing complexes, factories, livestock sites, construction sites, shopping malls, offices, roads, and stations may be classified as objects other than designated objects. Of course, eventually, objects in these segments may also correspond to thing objects. For example, in order to determine an event for an existing object, it is distinct from determining a specific tool or detecting an object possessed by a human object (eg, cigarettes, garbage, etc.) It can be seen as an object that is merely a background that is not particularly interested in poetry. The image processing apparatus 120 presents event types that can be used (or selected) for each classified segment. This may be preset as data in advance, and the user may select a plurality of events to be detected from the suggested event types.

For example, in the image processing apparatus 120 according to an embodiment of the present invention, when a photographed image is received after the photographing apparatus 100 is installed in a specific place, the image processing apparatus 120 determines an event based on a user-specified object and automatically classifies the segment. operation can be performed using In the received image, if the background image as above, i.e., a factory or construction site, is included in the background area, the area is segmented and classified, and the types of events that can be set related to the segment are presented to users such as control personnel for example. receive For example, there may be many types of events that may occur in relation to a designated object at a construction site. Accordingly, the user can select only an event to be detected among them. For example, at a construction site, a fall accident of a worker may be detected as an event, and fire detection may be detected as an event. Accordingly, the user may not be very interested in fire detection, but may be only interested in a fall accident. Therefore, it can be seen that the image processing apparatus 120 finally detects an event related to a fall accident of a designated object at a construction site. On the other hand, with respect to a specific object, different segments in the same image may include roads. Accordingly, the road may operate to detect other events in relation to the specified object. On the road, a traffic accident of a specified object may be detected as an event, and as such, the image processing apparatus 120 detects different types of events according to the types of surrounding objects segmented around the specified object. it can be

Of course, the opposite is also possible. For example, a received captured image is segmented for a specific target, an event type and event occurrence condition (eg, object type, etc.) are set for each segment, and an event is detected based on this. For example, the image processing apparatus 120 allows the user to select N events to be detected in the image, and sets the object type, size, holding time (or time information), etc. as event occurrence conditions, and according to the setting The image processing apparatus 120 may start or perform an event detection operation. Of course, since such a setting operation can be set automatically through an artificial intelligence program or the like, the embodiment of the present invention will not be particularly limited to any one form. In other words, many events that can be set by workers and construction sites can be monitored only for specific events. Accordingly, depending on the place where the photographing apparatus 100 is installed, events that can be set in relation to these surrounding objects may be automatically set. In short, the events that can occur at a housing construction site and a large-scale construction site are different. Therefore, it can be automatically set based on these characteristics. This is because, in the former case, fire accidents are emphasized, and in the latter case, fall accidents of workers can be emphasized.

In the image processing apparatus 120 according to an embodiment of the present invention, when segment classification and event setting for each segment are completed, that is, when event setting for each segment is completed, whether manual setting or automatic setting, an event determined based on the tracking result of a specified object Various additional actions may be further performed, such as re-judging an event by comprehensively judging the situation of surrounding objects or re-selecting or adjusting the priorities of events occurring for each segment. For example, after generating big data by collecting metadata related to a designated object, an additional operation such as adjusting a priority based on the big data may be performed.

As mentioned above, in the prior art, a human object and an object object are detected in a received video frame, and an event is generated according to a pre-stored rule policy based on the interrelationship between the detected objects, or in some cases, the shooting area. It was common to set a region of interest (ROI), typically set a crosswalk as a region of interest, and generate an event when an event, such as a traffic accident, occurs in the region. However, in the embodiment of the present invention, an event of interest can be set by presenting an event suitable for a segment such as a specific place in the background image, and a simple setting operation can be performed even if there are many types of events, and the ROI is set as before. It is possible to recognize the complex situation in which the event occurred because the entire area is analyzed without doing so. In addition, by making metadata into big data, false positives and false positives in event detection can be improved. When learning big data through artificial intelligence (AI) programs, etc., the accuracy increases, which guarantees objectivity and deriving events based on 100 data rather than deriving results based on 10 data. It can be seen that it is based on the logic that accuracy can be increased.

Although the image processing apparatus 120 according to the embodiment of the present invention has been briefly mentioned above, the image processing apparatus 120 sets an ROI, determines an event based on this, segments a non-interest region other than the ROI, and finally generates an event. When making a decision, it can be applied to a method of generating an event by additionally considering the segmented surrounding objects. Compared with the previous operation, there may be differences in terms of setting a specific object, which is an object of interest, or setting a region of interest, but it is common in that a surrounding object, that is, a segmented region, such as a place, can be additionally considered. can Accordingly, in the embodiment of the present invention, it will not be particularly limited to any one method, such as whether to set the region of interest or whether to designate a designated object.

FIG. 2 is a block diagram illustrating a detailed structure of the image processing apparatus of FIG. 1 , and FIG. 3 is a block diagram illustrating a detailed structure of the segment processing unit of FIG. 2 .

As shown in FIG. 2 , the image processing apparatus 120 of FIG. 1 according to an embodiment of the present invention includes a communication interface unit 200 , a control unit 210 , a segment processing unit 220 , and a portion of the storage unit 230 . or all inclusive.

Here, “including some or all” means that some components such as the storage unit 230 are omitted to configure the image processing apparatus 120 , or some components such as the segment processing unit 220 are included in the control unit 210 . It means that it can be configured by being integrated with other components, such as, and will be described as including all in order to help a sufficient understanding of the invention.

The communication interface 200 communicates with the photographing apparatus 100 via the communication network 110 of FIG. 1 . In the process of performing communication, the communication interface unit 200 may perform operations such as modulation/demodulation, and may perform operations such as encoding/decoding and muxing/demuxing. For example, the photographing apparatus 100 may generate metadata related to a photographing object in an image when photographing an arbitrary space or an arbitrary place, and such metadata may be transmitted together with the photographed image. In general, in the case of a 3D camera, meta data can be generated and transmitted, so the embodiment of the present invention will not be limited to any one form.

The control unit 210 is in charge of overall control operations of the communication interface unit 200, the segment processing unit 220, and the storage unit 230 of FIG. 2 constituting the image processing apparatus 120 of FIG. 1 . Typically, the control unit 210 may store the captured image provided from the communication interface unit 200 in the storage unit 230 , then retrieve it and provide it to the segment processing unit 220 . In addition, when the segment processing unit 220 generates and provides the analysis result of the captured image in a specified format, the control unit 210 systematically classifies it and stores it in the DB 120a of FIG. 1 .

The segment processing unit 220 may perform, for example, an analysis operation of the captured image. Above all, the segment processing unit 220 may include a plurality of areas segmented around a designated object or an object of interest according to an embodiment of the present invention, and of course, the corresponding area may include a place related to an event or a building of the place. , it is possible to determine the event set for each segment. For example, when a designated object for object tracking is set, the segment processing unit 220 automatically classifies a segment including a surrounding object based on this, and presents an event type that can be used for each classified segment to a user or an administrator. Thus, N events to be detected in the image can be selected. This has already been described above. Accordingly, the segment processing unit 220 may determine an event determined by comprehensively considering the specified object and the surrounding segmented surrounding objects.

More specifically, the segment processing unit 220 may be the entire image in the background region of the specified object, but if the background image is based on the background image, the segment processing unit 220 segments the foreground, that is, fragments or splits a plurality of regions into each region, and classifies it to segment Executes event setting operation for each event. For example, if two segments are analyzed in the captured image, an event may be set for each segment. As described above, the segment processing unit 220 selects N events to be detected by presenting the types of available events for each segment. And set the event occurrence condition additionally. It is to enable additional setting of event occurrence conditions after setting the event type. The event occurrence condition may include object type, size, and holding time. When an event type and occurrence condition in consideration of such a segment are set, event detection can be performed based on this.

As a result of the above, the image processing apparatus 120 according to an embodiment of the present invention can present an event suitable for a segment, and can simply perform a setting operation even if there are many types of events, without setting an ROI as in the past. By analyzing the entire domain, it may be possible to recognize the complex circumstances in which the event occurred. In addition, when meta data is generated as big data according to the above process and is used, it is possible to improve false positives or false positives, that is, non-detection of events. Here, if the metadata reflects the properties of a specified object, it can be considered that big data is generated by collecting metadata according to time change.

The segment processing unit 220 of FIG. 2 may have, for example, the configuration of the segment processing unit 220 ′ as in FIG. 3 . Of course, the configuration may be made in various ways, and this configuration may be configured by a S/W module, a H/W module, or a combination thereof. In the embodiment of the present invention, it will not be particularly limited to any one form in such a configuration.

3 , the image analysis unit 300 may analyze the received captured image and output an analysis result. For example, an operation of detecting and classifying a specified object in a video frame and tracking the object may be performed. An object detector and an object classifier may be included for object detection and classification, and DCNN object detection and classification may be performed for such object detection and classification. Accordingly, erroneous detection of an object, etc. may be reduced. Object tracking can be viewed as tracking the movement of a specific object in multiple video frames. Of course, the image analyzer 300 may perform automatic segment classification on the background image in addition to object detection in the process of image analysis. For example, the foreground image may be divided into a plurality of regions in a matrix form, that is, segmented, and segmented by determining whether a predetermined specific object is included during segmentation. The target of the segmentation may be limited to a construction site, a road, a parking lot, etc. as preset. The object of segmentation may be a place. In other words, segmentation can be used to ensure that events that fit the segment are detected when determining the events of a given object. Although the segmentation of the foreground image or the background image has been exemplified above, since it is possible to segment the entire captured image, the embodiment of the present invention will not be limited to any one form.

Also, the segment setting unit 310 performs a setting operation performed in relation to a segment by, for example, a user, an administrator, or a programmer. For example, the image analysis unit 300 may set N events to be detected in the image by presenting the type of event to the user for the automatically classified segment. In addition, when N events are selected, event occurrence conditions such as object type or size may be set.

The event determination unit 320 may determine an event based on the segment setting information set by the segment setting unit 310 and based on the analysis result provided by the image analysis unit 300 . In the image analysis unit 300 , if object detection, classification, and tracking are performed for a plurality of objects, the event determination unit 320 determines an event for each segment as set by the user or the like in the segment setting unit 310 . can see. In other words, if the segment area is a construction site, an event set in this regard is detected. In addition, if it is a parking lot, an event set in this regard is detected. For example, since the event type is set and the object type is also set, it is possible to detect an event only for the operator, not the supervisor, at the construction site.

Also, when an object of interest is designated as in another embodiment of the present invention, the event determining unit 320 determines an event based on a rule for the designated object, and based on the setting information of the segment setting unit 310 , It is also possible to reconfirm the event judged by In this process, the event determination unit 320 may perform an operation such as adjusting the priority of the event occurring for each segment, that is, reselecting it. In short, the event determination unit 320 determines the event A based on the tracking result of the specified object, but may determine the event B as the event B in consideration of the segment, that is, the surrounding situation comprehensively. Above all, since the embodiment of the present invention determines or comprehensively determines an event for a specified object through segmentation of a portion such as a place in the entire image or background area, it will not be limited to any one form.

Meanwhile, the storage unit 230 of FIG. 2 may sequentially store the captured images provided under the control of the control unit 210 in a memory and then retrieve them and provide them to the segment processing unit 220 . In addition, the storage unit 230 temporarily stores the image analysis result in a specified format including the segment data processed by the segment processing unit 220 and systematically classifies it in the DB 120a of FIG. 1 under the control of the control unit 210 . You can print it out to be saved.

Meanwhile, as another embodiment of the present invention, the control unit 210 may include a CPU and a memory, and may be formed as a single chip. The CPU includes a control circuit, an arithmetic unit (ALU), a command interpreter and a registry, and the memory may include a RAM. The control circuit performs a control operation, the operation unit performs an operation operation of binary bit information, and the instruction interpretation unit converts a high-level language into a machine language and a machine language into a high-level language, including an interpreter or compiler. , the registry may be involved in software data storage. According to the above configuration, for example, at the beginning of the operation of the image processing apparatus 120, a program stored in the segment processing unit 220 is copied, loaded into a memory, that is, a RAM, and then executed, thereby rapidly increasing the data operation processing speed. can do it

4 is a flowchart illustrating a driving process of an image processing apparatus according to an embodiment of the present invention.

Referring to FIG. 4 together with FIG. 1 for convenience of explanation, the image processing apparatus 120 according to an embodiment of the present invention receives a captured image in an arbitrary space (S400). Of course, here, the captured image is not necessarily an image provided by a camera, etc., but may be an image stored in a storage medium such as USB or a server such as a DVR and provided.

In addition, the image processing apparatus 120 automatically classifies by segmenting a plurality of areas including designated different types of objects (eg, construction sites, parking lots, etc.) including places in the received captured image into each area, An event for each segment may be processed based on the automatically classified segment (S410).

For example, as an embodiment, the image processing apparatus 120 determines an event based on a rule based on a specified object or based on deep learning of artificial intelligence, and then considers the segment or sets setting information related to the segment. An operation such as re-adjusting the event determined based on the basis may be performed. Accordingly, since the event is determined in consideration of the surrounding circumstances for the designated object, the accuracy of event determination may be increased.

Of course, the image processing apparatus 120 according to the embodiment of the present invention performs segmentation on the entire image or background as another embodiment as described above, and then determines the type of event set for each segment and the event occurrence condition. Since the event may be detected based on the basis, the embodiment of the present invention will not be particularly limited to any one operation form. In other words, since the user has set the event type and event occurrence condition that he wants to check what event in the housing complex and what event he wants to check in relation to the station, the image processing device 120 performs image analysis and segment processing accordingly. After detecting an event by performing it, the event may be notified when the corresponding event is detected.

In addition to the above, the image processing apparatus 120 according to an embodiment of the present invention can perform various operations, and since other details have been sufficiently described above, those contents will be replaced.

5 is a flowchart illustrating a driving process of an image processing apparatus according to another embodiment of the present invention.

For convenience of explanation, referring to FIG. 5 together with FIG. 1 , the image processing apparatus 120 according to another embodiment of the present invention performs an operation for setting an event for each segment based on the captured image received from the photographing apparatus 100 . It can be performed (S500 ~ S525).

For example, when a captured image is received, the image processing apparatus 120 may perform segmentation of a foreground image or a background area, although it may be the entire image, and then automatically classify the image and set an event for each segment. For example, when a segment is a construction site, if the system presents the available event types for each segment programmatically, the user can select N events to detect and set the event occurrence conditions in relation to the selected event. In addition, in the case of an office, the system, that is, a pre-installed program may select an event of interest by presenting a selectable event type related to the office to the user, and may set the event occurrence condition as described above.

As described above, when the setting operation for each segment is completed, the image processing apparatus 120 may detect an event based on the setting information (S530 to S560). In other words, if there are a plurality of segments in which events are set in the received captured image, the image processing apparatus 120 according to an embodiment of the present invention may simultaneously detect and notify the plurality of events.

In addition, as an embodiment, the image processing apparatus 120 determines an event based on rule-based or artificial intelligence deep learning through object detection, classification, and tracking with respect to a designated object of interest to the user, and then, as above, An event can be finally determined based on information set by the user for each segment. For example, if an event is detected and notified, but the event is not interested in the setting information set by the user in relation to the surrounding segment, the notification of the corresponding event may be cancelled.

Furthermore, the image processing apparatus 120 may generate big data by collecting metadata related to an object in which an event has occurred based on time change. Accordingly, big data can be used for event determination ( S570 ), and after processing the metadata into big data, operations such as re-selection of priorities of events occurring for each segment may be performed. For example, the list of events may be ranked based on statistics at a specific point in time based on big data. And, if time changes, the corresponding ranking can be adjusted based on big data.

In addition to the above, the image processing apparatus 120 according to another embodiment of the present invention can perform various operations, and since other details have been sufficiently described above, those contents will be replaced.

On the other hand, even though it has been described that all components constituting the embodiment of the present invention are combined or operated in combination, the present invention is not necessarily limited to this embodiment. That is, within the scope of the object of the present invention, all the components may operate by selectively combining one or more. In addition, all of the components may be implemented as one independent hardware, but a part or all of each component is selectively combined to perform some or all of the combined functions in one or a plurality of hardware program modules It may be implemented as a computer program having Codes and code segments constituting the computer program can be easily inferred by those skilled in the art of the present invention. Such a computer program is stored in a computer-readable non-transitory computer readable media, read and executed by the computer, thereby implementing an embodiment of the present invention.

Here, the non-transitory readable recording medium refers to a medium that stores data semi-permanently and can be read by a device, not a medium that stores data for a short moment, such as a register, cache, memory, etc. . Specifically, the above-described programs may be provided by being stored in a non-transitory readable recording medium such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM, and the like.

In the above, preferred embodiments of the present invention have been illustrated and described, but the present invention is not limited to the specific embodiments described above, and it is common in the technical field to which the present invention pertains without departing from the gist of the present invention as claimed in the claims. Various modifications may be made by those having the knowledge of, of course, and these modifications should not be individually understood from the technical spirit or perspective of the present invention.

100: photographing device 110: communication network
120: image processing device 200: communication interface unit
210: control unit 220, 220': segment processing unit
230: storage unit 300: image analysis unit
310: segment setting unit 320: event determination unit

Claims (8)

a communication interface unit for receiving a photographed image of an arbitrary space; and
Segmentation of a plurality of regions including objects of different types of people and places in the received captured image into regions that do not include the specified object based on the determination result by determining whether a preset specified object is included ( segment) and automatically classifies the segment when the setting for automatic classification performed by a user, manager, or programmer in relation to the segment is completed, and sets the automatically classified event for each segment to the event type set by the user, manager, or programmer and a control unit that determines and processes the event based on preset information of the event occurrence condition;
The controller determines an event for the designated object preset by the user in relation to the arbitrary space based on the captured image, segments a background image excluding the designated object into each region, and records the determined event Image processing device for automatic segment classification that judges by segment. delete According to claim 1,
The control unit detects at least one of a housing complex, a factory, a livestock site, a construction site, a shopping mall, an office, a road, and a station as the different types of objects other than the designated object and performs the segmentation automatically. image processing device for According to claim 1,
The control unit is an image processing apparatus for automatic segment classification for re-judging a complex situation based on the automatically classified segment with respect to the event determined in relation to the specified object. According to claim 1,
The control unit presents an event type that can be selected for each automatically classified segment to the user, receives an event that the user wants to detect and an event occurrence condition related to the segment, respectively, and performs an event determination operation for each segment Image processing device for classification. 6. The method of claim 5,
The control unit is an image processing apparatus for automatic segment classification that receives the type, size, and time information of an object as the event occurrence conditions, respectively. 6. The method of claim 5,
The control unit prioritizes the events occurring for each segment using metadata generated including attribute information related to the specified object in the captured image and surrounding objects of the specified object and big data generated by collecting the metadata Image processing device for automatic segment classification that adjusts the ranking. A method of driving an image processing apparatus including a communication interface unit and a control unit, the method comprising:
receiving, by the communication interface unit, a photographed image of an arbitrary space; and
The control unit determines whether a plurality of regions including objects of different types of people and places in the received captured image include a preset designated object, and each region that does not include the designated object based on the determination result and automatically classifies the segment when the setting for automatic classification performed by a user, administrator, or programmer in relation to the segment is completed, and sets the automatically classified event by segment to an event set by the user, administrator, or programmer Determining and processing based on the preset event information of the type and event occurrence condition;
The control unit determines an event for the specified object preset by the user in relation to the arbitrary space based on the captured image, segments a background image excluding the specified object into each region, and records the determined event judging by segment;
A method of driving an image processing device for automatic segment classification, including

Images