Artificial intelligence image analysis-based automatic camera configuration

Abstract

The present specification relates to an artificial intelligence image analysis-based automatic camera configuration system, an automatic camera configuration device, and a method for automatic camera configuration using same. The artificial intelligence image analysis-based automatic camera configuration system includes: a first surveillance camera which uses a wide-angle lens to capture a wide-area first surveillance area, thereby acquiring a first surveillance image; and a second surveillance camera having a PTZ function for obtaining a second surveillance image by capturing a second surveillance area within the first surveillance area. The first surveillance camera analyzes the first surveillance image on the basis of a first object analysis artificial intelligence model to derive an object feature analysis result for a detected object, and generates a profile for configuring the second surveillance camera on the basis of the object feature analysis result. The second surveillance camera controls PTZ according to the profile and can configure the camera.

Description

Automatic camera settings based on AI image analysis

[0001] This specification is about the camera settings.

[0002] In monitoring by filming the surveillance area, a fisheye camera with a fisheye lens that has a 360-degree surveillance range is used to monitor a wide surveillance area, and at the same time, a PTZ (Pan-Tilt-Zoom) camera with a pan (horizontal rotation) (Pan), tilt (tilt) (Tilt), and zoom (zoom) adjustment function that can film while changing the surveillance range in a variety of ways, although it has a relatively narrow surveillance range, is used to freely change the surveillance area and monitor.

[0003] Fisheye cameras have a wide surveillance area and can reduce blind spots, but it is difficult to obtain detailed surveillance data. PTZ cameras allow free adjustment of the surveillance area, but the surveillance area must be set by individually adjusting the pan, tilt, and zoom.

[0004] The above-described content is only intended to help understand the background technology for the technical ideas of the present invention, and therefore cannot be understood as content corresponding to prior art known to those skilled in the art in the technical field of the present invention.

[0005] An embodiment of the present specification aims to provide a camera setting technology using object recognition artificial intelligence that automatically sets a camera to obtain detailed surveillance images of detected objects by analyzing surveillance images.

[0006] In addition, one embodiment of the present specification aims to provide a technology that can automatically set the PTZ function of a PTZ camera so as to obtain a detailed image related to a recognized object by using the image analysis results of a wide-angle camera having an artificial intelligence-based object recognition function.

[0007] The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0008] The present specification proposes an automatic camera setting system based on artificial intelligence image analysis. The automatic camera setting system based on artificial intelligence image analysis includes a first surveillance camera having a wide-angle lens for photographing a first surveillance area in a wide area to obtain a first surveillance image; and a second surveillance camera having a PTZ function for photographing a second surveillance area within the first surveillance area to obtain a second surveillance image; wherein the first surveillance camera analyzes the first surveillance image based on a first object analysis artificial intelligence model to derive an object feature analysis result for a detected object, and generates a profile for setting the second surveillance camera based on the object feature analysis result, and the second surveillance camera can control the PTZ and set the camera according to the profile.

[0009] The above artificial intelligence image analysis-based camera automatic setting system and other embodiments may include the following features.

[0010] According to an embodiment, the first object analysis artificial intelligence model may be trained to recognize an object in an input image and derive an object feature analysis result including an object category, a heat map of the object, an average movement speed of the object, and location information of the object as an object feature analysis result for the recognized object.

[0011] In addition, according to an embodiment, the first surveillance camera may be characterized in that, when a pre-designated object is detected within the first surveillance area as a result of analyzing the first surveillance image, the second surveillance camera generates a profile to obtain an image of a certain area including the area where the object is detected.

[0012] In some embodiments, the first surveillance camera may also generate the profile based on a detection frequency of the pre-specified object or a category of the pre-specified object.

[0013] In addition, according to an embodiment, the first surveillance camera may generate a profile for the PTZ settings of the second surveillance camera to calculate the moving speed of the pre-designated object if the category of the pre-designated object is a vehicle, and track the pre-designated object according to the calculated moving speed.

[0014] In an embodiment, the first surveillance camera may also generate a profile with the additional use of a wide dynamic range (WDR) function if the category of the pre-specified object is a vehicle.

[0015] In an embodiment, further, the second surveillance camera may include a second object analysis artificial intelligence model trained to recognize an object in an input image, and may analyze the second surveillance image using the second object analysis artificial intelligence model, and when a pre-designated object is identified within the second surveillance area, may instruct the first surveillance camera to track and store the movement path of the pre-designated object within the first surveillance area.

[0016] In an embodiment, the first surveillance camera may also recommend a preset profile based on a category of an object detected within the first surveillance area as a result of analyzing the first surveillance image.

[0017] Meanwhile, the present specification proposes an automatic configuration camera based on artificial intelligence image analysis. The automatic configuration camera based on artificial intelligence image analysis includes a camera module for photographing a wide surveillance area; and an image analysis module for analyzing a surveillance image of the surveillance area acquired by the camera module based on an object analysis artificial intelligence model, producing an object feature analysis result for a detected object, and generating a profile for configuring the camera module based on the object feature analysis result; wherein the image analysis module can configure the camera module according to the profile.

[0018] The above artificial intelligence image analysis-based automatic setting camera and other embodiments may include the following features.

[0019] According to an embodiment, the image analysis module, when an object is recognized in some area within the surveillance area as a result of analyzing the surveillance image for a predetermined period of time, assigns a weight to each area in which the object is recognized based on at least one of the detection frequency of the object and the movement of the object, generates a profile that sets the image quality acquired differently for each area in which the object is recognized based on the weight, and sets the camera module according to the profile.

[0020] In some embodiments, the image quality may also be determined by at least one of a quantization parameter, a bitrate, and an applied codec.

[0021] In addition, according to an embodiment, the image analysis module may, when an object is recognized within the surveillance area as a result of analyzing the surveillance image for a predetermined period of time, determine a time period in which the object is recognized based on at least one of the detection frequency of the object and the movement of the object, generate a profile that sets the image quality acquired by the camera module differently according to the time period, and set the camera module according to the profile.

[0022] In addition, according to an embodiment, the image analysis module may generate a profile that sets the image quality of a certain area including the recognized object differently from an area outside the certain area, and set the camera module according to the profile, if, as a result of analyzing the surveillance image, an object recognized within the surveillance area is identified as a pre-specified object or the movement of the recognized object exceeds a pre-specified standard.

[0023] On the other hand, the present specification proposes a camera automatic setting method by an artificial intelligence image analysis-based camera automatic setting system. The camera automatic setting method of an artificial intelligence image analysis-based camera automatic setting system including a first surveillance camera and a second surveillance camera may include: a step in which the first surveillance camera photographs a first surveillance area to obtain a first surveillance image of the first surveillance area; a step in which the first surveillance camera analyzes the first surveillance image based on a first object analysis artificial intelligence model to derive an object feature analysis result for a detected object; a step in which the first surveillance camera generates a profile for setting the second surveillance camera so that the second surveillance camera obtains a second surveillance image of a second surveillance area within the first surveillance area based on the object feature analysis result; and a step in which the second surveillance camera controls a PTZ according to the profile and sets the camera to obtain the second surveillance image.

[0024] The automatic camera setting method and other embodiments by the above artificial intelligence image analysis-based automatic camera setting system may include the following features.

[0025] According to an embodiment, the first object analysis artificial intelligence model may be trained to recognize an object in an input image and derive an object feature analysis result including an object category, a heat map of the object, an average movement speed of the object, and location information of the object as an object feature analysis result for the recognized object.

[0026] In an embodiment, further, the step of generating a profile for setting the second surveillance camera by the first surveillance camera may include a step of generating a profile so that, when a pre-specified object is detected within the first surveillance area as a result of analyzing the first surveillance image, the second surveillance camera acquires an image for a certain area including the area where the object is detected.

[0027] In an embodiment, furthermore, the step of generating a profile for setting the second surveillance camera by the first surveillance camera may include a step of generating the profile by the first surveillance camera further based on a detection frequency of the pre-specified object or a category of the pre-specified object.

[0028] In addition, according to an embodiment, the step of generating a profile for setting the second surveillance camera by the first surveillance camera may include the step of calculating a moving speed of the pre-specified object if the category of the pre-specified object is a vehicle, and generating a profile for the PTZ setting of the second surveillance camera to track the pre-specified object according to the calculated moving speed.

[0029] In an embodiment, furthermore, the step of generating a profile for setting the second surveillance camera by the first surveillance camera may include the step of generating a profile to which use of a wide-area backlight compensation function is added if the category of the pre-specified object is a vehicle.

[0030] In addition, according to an embodiment, the automatic camera setting method by the automatic camera setting system based on artificial intelligence image analysis may further include a step in which the second surveillance camera analyzes the second surveillance image using a second object analysis artificial intelligence model, the second surveillance camera includes a second object analysis artificial intelligence model learned to recognize an object in an input image; and a step in which, when a pre-designated object is identified within the second surveillance area, the second surveillance camera instructs the first surveillance camera to track and store a movement line of the pre-designated object within the first surveillance area.

[0031] The embodiments disclosed herein have the effect of providing a camera setting technology using object recognition artificial intelligence that automatically sets a camera to obtain detailed surveillance images of detected objects by analyzing surveillance images.

[0032] In addition, the embodiment disclosed in the present specification has the effect of automatically setting the PTZ camera PTZ function to obtain detailed images related to a recognized object using the image analysis results of a wide-angle camera having an artificial intelligence-based object recognition function.

[0033] Meanwhile, the effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0034] The following drawings attached to this specification illustrate preferred embodiments of the present invention and, together with specific details for carrying out the invention, serve to further understand the technical idea of ​​the present invention. Therefore, the present invention should not be interpreted as being limited to matters described in such drawings.

[0035] Figure 1 illustrates an example of an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0036] Figure 2 is a block diagram schematically showing the internal configuration of the wide-angle camera illustrated in Figure 1.

[0037] Figure 3 is a block diagram of the AI ​​processing unit of Figure 2.

[0038] Fig. 4 is a block diagram schematically showing the internal configuration of the PTZ camera illustrated in Fig. 1.

[0039] FIG. 5 is a drawing illustrating a camera automatic setting method by an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0040] FIG. 6 is a diagram illustrating a method for tracking an automobile object by an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0041] FIG. 7 is a diagram illustrating a case in which a wide-angle camera tracks the movement path of an object identified by a PTZ camera in an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0042] Figure 8 is a flowchart illustrating a camera automatic setting method based on artificial intelligence image analysis according to one embodiment.

[0043] The technology disclosed herein can be applied to automatic camera setup technology based on image analysis using artificial intelligence. However, the technology disclosed herein is not limited to this technology and can be applied to any device or method to which the technical principles of the technology can be applied.

[0044] It should be noted that the technical terms used in this specification are merely used to describe specific embodiments and are not intended to limit the scope of the technology disclosed herein. Furthermore, unless specifically defined otherwise herein, the technical terms used herein should be interpreted as having a meaning generally understood by a person of ordinary skill in the art to which the technology disclosed herein pertains, and should not be interpreted in an excessively broad or narrow sense. Furthermore, if a technical term used herein is an incorrect technical term that does not accurately express the scope of the technology disclosed herein, it should be replaced with a technical term that can be correctly understood by a person of ordinary skill in the art to which the technology disclosed herein pertains. Furthermore, general terms used herein should be interpreted according to their dictionary definitions or according to the context, and should not be interpreted in an excessively narrow sense.

[0045] While terms including ordinal numbers, such as "first" and "second," used herein may be used to describe various components, these components should not be limited by these terms. These terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, a first component could be referred to as a "second component," and similarly, a second component could also be referred to as a "first component."

[0046] Hereinafter, embodiments disclosed in this specification will be described in detail with reference to the attached drawings. Regardless of the drawing numbers, identical or similar components are given the same reference numbers and redundant descriptions thereof will be omitted.

[0047] Additionally, when describing the technology disclosed in this specification, detailed descriptions of related known technologies will be omitted if they are deemed to obscure the gist of the technology disclosed in this specification. Furthermore, it should be noted that the attached drawings are intended solely to facilitate understanding of the concepts of the technology disclosed in this specification and should not be construed as limiting the scope of the technology.

[0048] An image or video according to one embodiment of the present invention includes both still images and moving images unless there is a special limitation.

[0049] Throughout the specification, a device or terminal includes a communication terminal or communication device capable of wired or wireless communication with a server or other device. The form of the device or terminal may be various, such as a mobile phone, smartphone, smart pad, laptop computer, desktop computer, smart TV, wearable device, mirror-type display device, smart mirror, etc. Wearable devices may be various, such as watch-type terminals, glass-type terminals, HMD, etc. Furthermore, the terminal is not limited to these forms and may be implemented as various electronic devices.

[0050] Hereinafter, embodiments are described in detail with reference to the attached drawings.

[0051] Figure 1 illustrates an example of an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0052] Referring to FIG. 1, an artificial intelligence image analysis-based camera automatic setting system (1000) according to an embodiment may be configured to include a wide-angle camera (100) and a PTZ camera (200).

[0053] The wide-angle camera (100) may be a camera equipped with a wide-angle lens to capture a surveillance area with a wide angle of view. The wide-angle camera (100) may have a fixed angle of view. The wide-angle lens may be a fisheye lens with a 360-degree angle of view. This can capture a surveillance area with a 360-degree angle of view to obtain an image of the surveillance area. In the present embodiment, a camera using a fisheye lens as the wide-angle lens is used as an example, but the present invention is not limited to a fisheye camera, and any camera with a much larger angle of view than the PTZ camera (200) may be used depending on the embodiment.

[0054] The PTZ camera (200) is a camera with pan, tilt, and zoom adjustment functions, and acquires an image of the surveillance area by capturing a detailed surveillance area with a narrower angle of view than the wide-angle camera (100).

[0055] The wide-angle camera (100) can detect and recognize objects using an artificial intelligence-based object analysis model in an image of an acquired surveillance area, and based on the result, can generate a profile regarding settings for adjusting the pan, tilt, and zoom of the PTZ camera (200) to acquire detailed images of the area where the object is detected, i.e., the event area.

[0056] Since the wide-angle camera (100) knows its own field of view, i.e., the position of the PTZ camera (200) within its surveillance area, it can calculate setting values ​​for driving control of the pan, tilt, and zoom of the PTZ camera (200) to track a specific object detected within its surveillance area or to obtain a detailed surveillance image of a specific object.

[0057] The PTZ camera (200) receives a profile or preset for camera setting values ​​for adjusting pan, tilt, and zoom generated by the wide-angle camera (100) from the wide-angle camera (100) and operates pan, tilt, and zoom, thereby obtaining a detailed surveillance image of the event area.

[0058] The wide-angle camera (100) can use the artificial intelligence-based object analysis model to create a heat map for each object category, calculate the average moving speed of the object, calculate the location of the object, etc., and build a database using the generated data. After a certain period of time, the wide-angle camera (100) can transmit a customized recommended PTZ setting for each object category to the PTZ camera (200) based on the built database, thereby enabling the PTZ camera (200) to obtain a detailed image of the object. The PTZ camera (200) can move to the corresponding coordinates according to the recommended PTZ setting value, adjust the angle of view, and then automatically set it to the received recommended setting value.

[0059] In this embodiment, the wide-angle camera (100) and the PTZ camera (200) are shown as being separated from each other, but depending on the embodiment, the wide-angle camera (100) and the PTZ camera (200) may be provided in the same housing and implemented as a single device.

[0060] The system (1000) may further include a video management device (300). The video management device (300) may include a video security solution such as a DVR, CMS, NVR, VMS, or a display device. The video management device (300) may receive video data captured by a wide-angle camera (100) and a PTZ camera (200) through a network, output the data to an administrator through a display, and store the received video data.

[0061] Figure 2 is a block diagram schematically showing the internal configuration of the wide-angle camera illustrated in Figure 1.

[0062] Referring to FIG. 2, a wide-angle camera (100) may be configured to include an image sensor (110), an encoder (120), a memory (130), a communication unit (140), an AI processing unit (150), and a control unit (160).

[0063] The image sensor (110) performs the function of capturing a surveillance area and obtaining a surveillance image, and can be implemented as, for example, a CCD (Charge-Coupled Device) sensor, a CMOS (Complementary Metal-Oxide-Semiconductor) sensor, etc. The image sensor (110) can obtain a 360-degree image through a fish-eye lens (not shown) located in front of it.

[0064] The encoder (120) performs an operation of encoding a surveillance image acquired through an image sensor (110) into a digital signal, which may follow, for example, H.264, H.265, MPEG (Moving Picture Experts Group), M-JPEG (Motion Joint Photographic Experts Group) standards, etc.

[0065] The memory (130) can store a program for the operation of the control unit (160) and temporarily store input / output data and generated data. In addition, the memory (130) can store video data, audio data, still images, metadata, etc. The metadata may be data including object detection information (movement, sound, intrusion into a designated area, etc.) captured in the surveillance area, object identification information (person, car, face, hat, clothing, etc.), and detected location information (coordinates, size, etc.).

[0066] In addition, the still image is generated together with the metadata and stored in the memory (130), and can be generated by capturing image information for a specific analysis area among the image analysis information. For example, the still image can be implemented as a JPEG image file. For example, the still image can be generated by cropping a specific area of ​​the image data determined to be an identifiable object among the image data of the surveillance area detected in a specific area and for a specific period of time, and this can be transmitted in real time together with the metadata.

[0067] The communication unit (140) can transmit the video data, audio data, still images, and / or metadata to the video management device (300 of FIG. 1). According to one embodiment, the communication unit (140) can transmit the video data, audio data, still images, and / or metadata to the video management device in real time. The communication unit (140) can perform at least one communication function among wired / wireless Local Area Network (LAN), Wi-Fi, ZigBee, Bluetooth, and Near Field Communication.

[0068] The AI ​​processing unit (150) is for processing images based on artificial intelligence, and performs object detection, object identification, and object tracking algorithms based on deep learning learned from images acquired through the wide-angle camera (1000) of the AI ​​image analysis-based camera automatic setting system (1000) according to one embodiment of the present specification. The AI ​​processing unit (150) may be implemented as one module with the control unit (160) that controls the entire system, or may be implemented as an independent module. The AI ​​processing unit (150) may also be implemented as a separate module or device separated from the wide-angle camera (100). In addition, the control unit (160) may include the AI ​​processing unit (150) to configure an image analysis module to be described later.

[0069] Figure 3 is a block diagram of the AI ​​processing unit of Figure 2.

[0070] The AI ​​processing unit (150) may include an electronic device including an AI module capable of performing AI processing, or a server including the AI ​​module. In addition, the AI ​​processing unit (150) may be included as part of at least a portion of the wide-angle camera (100) illustrated in FIG. 2, and may be configured to perform at least a portion of the AI ​​processing performed by the wide-angle camera.

[0071] The AI ​​processing of the AI ​​processing unit (150) may include all operations related to the control of the control unit (160) illustrated in FIG. 2 and all operations for image recognition through artificial intelligence learning. For example, the AI ​​processing unit (150) may perform an operation to recognize the appearance of a vehicle or person within the surveillance area by performing AI processing on the zero data for the surveillance area acquired by the image sensor (110) of the wide-angle camera (100) based on the YOLO model. The AI ​​processing unit (150) may also be included as a component of the image processing unit (231) illustrated in FIG. 4.

[0072] The above AI processing unit (150) may include an AI processor (151), a memory (155), and / or a communication unit (157).

[0073] The above AI processing unit (150) is a computing device capable of learning a neural network, and can be implemented as various electronic devices such as a server, desktop PC, notebook PC, tablet PC, etc., or can be implemented as a single chip.

[0074] The AI ​​processor (151) can learn a neural network using a program stored in the memory (155). In particular, the AI ​​processor (151) can learn a neural network for recognizing device-related data. Here, the neural network for recognizing device-related data can be designed to simulate the human brain structure on a computer, and can include a plurality of network nodes having weights that simulate neurons of a human neural network. The plurality of network modes can each exchange data according to a connection relationship so as to simulate the synaptic activity of neurons that exchange signals through synapses. Here, the neural network can include a deep learning model developed from a neural network model. In the deep learning model, a plurality of network nodes can be located in different layers and exchange data according to a convolution connection relationship. Examples of neural network models include various deep learning techniques such as deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), restricted Boltzmann machines (RBMs), deep belief networks (DBNs), and deep Q-networks, and can be applied to fields such as computer vision (CV), speech recognition, natural language processing, and speech / signal processing.

[0075] Meanwhile, the processor performing the functions described above may be a general-purpose processor (e.g., CPU), but may also be an AI-specific processor for artificial intelligence learning (e.g., GPU).

[0076] The memory (155) can store various programs and data required for the operation of the AI ​​processing unit (150). The memory (155) can be implemented as a non-volatile memory, a volatile memory, a flash memory, a hard disk drive (HDD), a solid state drive (SDD), etc. The memory (155) is accessed by the AI ​​processor (151), and data reading / writing / modifying / deleting / updating, etc. can be performed by the AI ​​processor (151). In addition, the memory (155) can store a neural network model (e.g., a deep learning model (156)) generated through a learning algorithm for data classification / recognition according to one embodiment of the present invention.

[0077] Meanwhile, the AI ​​processor (151) may include a data learning unit (152) that learns a neural network for data classification / recognition. The data learning unit (152) may learn criteria regarding which learning data to use to determine data classification / recognition and how to classify and recognize data using the learning data. The data learning unit (152) may acquire learning data to be used for learning and apply the acquired learning data to the deep learning model, thereby learning the deep learning model.

[0078] The data learning unit (152) may be manufactured in the form of at least one hardware chip and mounted on the AI ​​processing unit (150). For example, the data learning unit (152) may be manufactured in the form of a dedicated hardware chip for artificial intelligence (AI), or may be manufactured as a part of a general-purpose processor (CPU) or a graphics processor (GPU) and mounted on the AI ​​processing unit (150). In addition, the data learning unit (152) may be implemented as a software module. When implemented as a software module (or a program module including instructions), the software module may be stored in a non-transitory computer readable medium that can be read by a computer. In this case, at least one software module may be provided by an operating system (OS) or an application (application program).

[0079] The data learning unit (152) may include a learning data acquisition unit (153) and a model learning unit (154).

[0080] The learning data acquisition unit (153) can acquire learning data required for a neural network model for classifying and recognizing data. For example, the learning data acquisition unit (153) can acquire image data and / or sample data for transmitters on road infrastructure for input into the neural network model as learning data.

[0081] The model learning unit (154) can use the acquired learning data to learn the neural network model to have a judgment criterion on how to classify a given data. At this time, the model learning unit (154) can train the neural network model through supervised learning that uses at least some of the learning data as a judgment criterion. Alternatively, the model learning unit (154) can train the neural network model through unsupervised learning that discovers a judgment criterion by learning on its own using the learning data without guidance. In addition, the model learning unit (154) can train the neural network model through reinforcement learning that uses feedback on whether the result of the situation judgment according to the learning is correct. In addition, the model learning unit (154) can train the neural network model using a learning algorithm including error back-propagation or gradient descent.

[0082] Once the neural network model is trained, the model training unit (154) can store the trained neural network model in memory. The model training unit (154) can also store the trained neural network model in the memory of a server connected to the AI ​​processing unit (150) via a wired or wireless network.

[0083] The data learning unit (152) may further include a learning data preprocessing unit (not shown) and a learning data selection unit (not shown) to improve the analysis results of the recognition model or to save resources or time required for creating the recognition model.

[0084] The learning data preprocessing unit can preprocess the acquired data so that it can be used for learning to determine the situation. For example, the learning data preprocessing unit can process the acquired data into a preset format so that the model learning unit (154) can use the acquired learning data for learning to recognize image data for the transmitter.

[0085] Additionally, the learning data selection unit can select data required for learning from among the learning data acquired by the learning data acquisition unit (153) or the learning data preprocessed by the preprocessing unit. The selected learning data can be provided to the model learning unit (154). For example, the learning data selection unit can select only data included in a specific field as learning data by recognizing a specific field among the data sets collected through the network.

[0086] Additionally, the data learning unit (152) may further include a model evaluation unit (not shown) to improve the analysis results of the neural network model.

[0087] The model evaluation unit inputs evaluation data into the neural network model, and if the analysis results output from the evaluation data do not satisfy a predetermined standard, it can cause the model learning unit (152) to relearn. In this case, the evaluation data may be predefined data for evaluating the recognition model. For example, the model evaluation unit can evaluate that the predetermined standard is not satisfied if the number or ratio of evaluation data with inaccurate analysis results among the analysis results of the learned recognition model for the evaluation data exceeds a preset threshold.

[0088] Meanwhile, the AI ​​processing unit (150) illustrated in FIG. 3 is functionally divided into an AI processor (151), a memory (155), a communication unit (157), etc., but it should be noted that the aforementioned components may be integrated into one module and referred to as an AI module or an artificial intelligence-based image analysis module.

[0089] Referring back to FIG. 2, the AI ​​processing unit (150) may include an object analysis artificial intelligence model. The object analysis artificial intelligence model may be trained to recognize objects in surveillance images acquired from the image sensor (110) and derive object feature analysis results for the recognized objects. The object feature analysis results may include, for example, a heat map of the object, an average moving speed of the object, and location information of the object by category of the object, which may be classified into vehicles, animals, people, and fixed objects.

[0090] The control unit (160) may be implemented as a processor, and the processor may be configured to process commands of a computer program by performing basic arithmetic, logic, and input / output operations.

[0091] The control unit (160) can generate a profile or preset for setting the pan, tilt, and zoom of the PTZ camera (200 in FIG. 1) based on the object feature analysis results derived from the object analysis artificial intelligence model of the AI ​​processing unit (150).

[0092] For example, if a specific object designated in advance is detected within a surveillance area as a result of analyzing a surveillance image acquired from an image sensor (110), the control unit (160) can generate a profile that sets pan, tilt, and zoom to cause the PTZ camera (200 in FIG. 1) to acquire a surveillance image for a certain area including the area where the object is detected.

[0093] Additionally, the control unit (160) may analyze the surveillance image acquired from the image sensor (110) and recommend a preset profile based on the category of the object detected within the surveillance area. For example, if the detected object is a dynamic object, the pan and tilt may be set so that the PTZ camera can acquire the image while tracking the dynamic object.

[0094] The control unit (160) may generate a profile further based on the detection frequency of the pre-specified object or the category of the object. For example, the control unit (160) may generate a profile for setting the pan, tilt, and zoom of the PTZ camera only when the pre-specified object is a specific person, but the detection frequency of the person within the surveillance area is above a specific value. For example, the control unit (160) may also generate a profile for setting the pan, tilt, and zoom of the PTZ camera only when vehicles are detected within the surveillance area.

[0095] Meanwhile, if the category of the pre-specified object is a vehicle, the control unit (160) may generate a profile for the pan, tilt, and zoom settings of the PTZ camera so that, when detecting a vehicle within the surveillance area, the moving speed of the detected vehicle can be calculated and the detected vehicle can be tracked according to the calculated moving speed. In addition, if the category of the pre-specified object is a vehicle, the control unit (160) may generate a profile with additional setting values ​​for driving the Wide Dynamic Range (WDR) function of the PTZ camera so that it may be difficult to identify vehicle license plate information or the faces of vehicle passengers due to ambient light reflecting on the vehicle window or lighting such as the headlights or taillights of the vehicle.

[0096] Fig. 4 is a block diagram schematically showing the internal configuration of the PTZ camera illustrated in Fig. 1.

[0097] Referring to FIG. 4, the PTZ camera (200) includes an image acquisition unit (210), a control unit (220), and a communication unit (230).

[0098] The image acquisition unit (210) acquires a real-time image signal. The image acquisition unit (210) may be composed of an image sensor such as a CCD (charge coupled device) or a CMOS (complementary metal oxide semiconductor), and a lens that transmits light to the image sensor. The image acquisition unit (210) may have a relatively narrower angle of view than the wide-angle camera (100 in FIG. 1) described above. The image acquisition unit (210) can acquire a surveillance image by photographing a surveillance area existing within the surveillance area according to the angle of view of the wide-angle camera.

[0099] The control unit (220) controls the overall operation of the PTZ camera (200) and may include an image processing unit (231), a memory (232), and a PTZ control unit (233).

[0100] The image processing unit (231) processes the real-time input image signal to generate a surveillance image. The surveillance image is an image captured in the surveillance area of ​​the corresponding camera, and may include predetermined thumbnail image information corresponding to the corresponding surveillance area and image information of an event captured in the surveillance area.

[0101] The memory (232) can perform the role of storing and managing image information generated by the image processing unit (231), camera-specific information, camera position information, etc.

[0102] The PTZ control unit (233) can change the PTZ coordinates of the camera according to the set preset information or the profile related to the PTZ settings. The preset information or the profile can be set by the administrator through, for example, the video management device (300 in FIG. 1), and the administrator can control the position, direction, zoom level, etc. of the camera by changing the PTZ coordinates using the preset information or profile of the camera. The PTZ control unit (233) can control the PTZ and set the camera according to the profile or preset information related to the PTZ settings received from the wide-angle camera (100 in FIG. 2). The profile can include a PTZ setting / control command according to the ONVIF (Open Network Video Interface Forum) standard.

[0103] The control unit (220) may be implemented as a processor, and the processor may be configured to process commands of a computer program by performing basic arithmetic, logic, and input / output operations. The commands may be provided to the processor, i.e., the control unit (220), by the memory (232). For example, the control unit (220) may be configured to execute commands received according to program code stored in a recording device such as the memory (232).

[0104] Here, the control unit (220) may be implemented to execute instructions according to the code of the operating system included in the memory (232) and at least one program code. At this time, the components within the control unit (220), i.e., the image processing unit (231) and the PTZ control unit (233), may be understood to express different functions performed by the control unit (220) according to the control commands provided by the program code stored in the memory (232).

[0105] The communication unit (230) is composed of a communication module and an antenna, and is connected to the control unit (220) to transmit information such as images from the image processing unit (231) to other cameras such as an image management device (300 in FIG. 1) or a wide-angle camera (100).

[0106] Meanwhile, the image processing unit (231) can perform image analysis (object detection, object identification, etc.) in the same manner as the AI ​​processing unit (150), including the AI ​​processing unit (150) described with reference to FIGS. 2 to 3. The image processing unit (231) analyzes the surveillance image of the surveillance area acquired by the image acquisition unit (210) to detect objects such as people and cars within the surveillance area, and distinguish and identify their type, color, characteristics, gender, etc. The image processing unit (231) may include an object analysis artificial intelligence model trained to recognize objects in the surveillance image acquired by the image acquisition unit (210). The image processing unit (231) can analyze the surveillance image acquired by the image acquisition unit (210) using the object analysis artificial intelligence model to detect and identify objects within the surveillance area. At this time, the control unit (220) of the PTZ camera (200) with excellent object detection and identification capabilities can instruct the wide-angle camera (100 in FIG. 2) to track and store the movement path of the object of interest in the surveillance area of ​​the wide-angle camera when an object of interest of interest is identified within the surveillance area of ​​the PTZ camera (200).

[0107] Meanwhile, the wide-angle camera (100 in FIG. 1) described above may also be used to supplement the auto-tracking function of the PTZ camera (200). For example, when the control unit (220) detects an object of interest to the manager within the surveillance area, it activates the auto-tracking function of the PTZ camera and transmits information about the detection location of the object to the wide-angle camera, thereby allowing the wide-angle camera to track the movement of the object of interest within a wide surveillance area. The wide-angle camera tracks the movement of the object of interest received within a wide field of view and transmits PTZ setting values ​​that enable tracking along the movement of the object of interest to the PTZ camera (200), thereby allowing the PTZ camera (200) to accurately track the object of interest even in a crowded situation due to multiple objects within the surveillance area.

[0108] The components illustrated in Figures 2 through 4 are not essential, and wide-angle cameras and PTZ cameras may be implemented with more or fewer components. These components may be implemented in hardware, software, or a combination of hardware and software.

[0109] FIG. 5 is a drawing illustrating a camera automatic setting method by an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0110] Referring to FIG. 5(a), the wide-angle camera (100) photographs the surveillance area (A) according to the angle of view to obtain a surveillance image for the surveillance area (A), and the PTZ camera (200) photographs the surveillance area (B1) according to the current PTZ settings to obtain a surveillance image for the surveillance area (B1). The wide-angle camera (100) analyzes the acquired surveillance image using an object analysis artificial intelligence model to detect objects (OB) within the surveillance area (A), and then derives feature analysis results for the detected objects (OB).

[0111] Referring to FIG. 5(b), the wide-angle camera (100) sets a surveillance area (B2) of a certain size including the detected objects (OB) as a surveillance area to be photographed by the PTZ camera (200) in order to obtain a detailed surveillance image of the detected objects (OB) based on the feature analysis result. The wide-angle camera (100) creates a profile related to PTZ settings that sets / controls the pan, tilt, and zoom of the PTZ camera (200) so that the PTZ camera (200) can photograph the surveillance area (B2), and then transmits the profile to the PTZ camera (200). The PTZ camera (200) sets the pan, tilt, and zoom according to the received profile, and obtains a surveillance image of the surveillance area (B2).

[0112] FIG. 6 is a diagram illustrating a method for tracking an automobile object by an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0113] Referring to FIG. 6, a wide-angle camera (100) photographs a surveillance area (A) according to an angle of view to obtain a surveillance image for the surveillance area (A). The wide-angle camera (100) analyzes the surveillance image using an object analysis artificial intelligence model to detect and identify a specific object. When the wide-angle camera (100) detects an automobile object (OB) in the surveillance area (A), the camera calculates the size, position, and moving speed of the detected automobile object (OB), and then derives setting values ​​for pan, tilt, and zoom values ​​for a PTZ camera (200) for tracking the automobile object (OB) based on the derived results, and generates a profile for setting the PTZ camera (200) based on the derived results and transmits the profile to the PTZ camera (200). The PTZ camera (200) adjusts its position so that it can capture an automobile object (OB) within the field of view of the wide-angle camera (100) according to the received profile, sets a surveillance area (B1), and then adjusts the PTZ according to the calculated movement speed to track the automobile object (OB) moving to the surveillance area (B2).

[0114] FIG. 7 is a diagram illustrating a case in which a wide-angle camera tracks the movement path of an object identified by a PTZ camera in an artificial intelligence image analysis-based camera automatic setting system according to one embodiment.

[0115] Referring to FIG. 7, when a PTZ camera (200) detects a specific object (S) within a surveillance area (B) within a surveillance area (A) according to the angle of view of a wide-angle camera (100), it commands the wide-angle camera (100) to track the object (S). The wide-angle camera (100) can track the object (S) within its surveillance area (A) according to the command of the PTZ camera (200) and store a path (T) according to the movement of the object (S).

[0116] Below, a method for automatically setting a wide-angle camera (100) using an artificial intelligence-based object analysis model of the wide-angle camera (100) is described.

[0117] Referring again to FIG. 2, the control unit (160) analyzes the surveillance image of the surveillance area acquired by the image sensor (110) based on the object analysis artificial intelligence model of the AI ​​processing unit (150), and calculates an object feature analysis result for an object detected in the surveillance area. In addition, the control unit (160) generates a profile for setting the camera module based on the calculated object feature analysis result, and sets the wide-angle camera (100) according to the generated profile.

[0118] The control unit (160) analyzes the surveillance video and, if an object is recognized in a certain area within the surveillance area during a predetermined period of time, assigns a weight to each area where the object is recognized based on at least one of the detection frequency of the object or the movement of the object, creates a profile that sets the image quality acquired for each area where the object is recognized differently according to the weight, and sets the camera according to the profile. The image quality-related profile may have setting values ​​related to a quantization parameter, a bit rate, and an applied codec.

[0119] For example, the control unit (160) can assign weights to areas where an object has passed and store the weights for those areas as a profile. Areas with low weights are likely to be areas with little object movement or surrounding background areas. Areas with little movement are likely to be areas with low importance, so a high quantization parameter can be applied for efficient image compression to reduce the amount of image data with high compression efficiency.

[0120] The control unit (160) may determine the time period in which the object is recognized based on at least one of the detection frequency of the object and the movement of the object, if an object is recognized within the surveillance area as a result of analyzing the surveillance video for a predetermined time period, and may generate a profile that sets the image quality acquired by the image sensor (110) differently according to the time period, and set the camera according to the profile.

[0121] For example, in the case of roads, the road usage rate may vary depending on the commuting time. In this case, the control unit (160) automatically recommends a profile so that the quality of the image acquired from the image sensor (110) is set high during times when the road usage rate is high, such as the morning commute or the afternoon commute, or so that the quality of the image acquired from the image sensor (110) is set low during times when the road usage rate is low, thereby allowing the administrator to use different image quality depending on the time period.

[0122] In addition, if the control unit (160) analyzes the surveillance image acquired from the image sensor (110) and identifies an object recognized within the surveillance area as a pre-specified object or the movement of the recognized object exceeds a pre-specified standard, the control unit (160) can generate a profile that sets the quality of a certain area including the recognized object high and sets the image quality of an area not including the recognized object low, and set the camera according to the profile.

[0123] Figure 8 is a flowchart illustrating a camera automatic setting method based on artificial intelligence image analysis according to one embodiment.

[0124] Referring to FIGS. 1 to 5 and 8, a camera automatic setting method of an artificial intelligence image analysis-based camera automatic setting system (1000) including a wide-angle camera (100) and a PTZ camera (200) can be performed including the following processes.

[0125] First, a wide-angle camera (100) photographs a first surveillance area to obtain a first surveillance image of the first surveillance area (S110).

[0126] Next, the wide-angle camera (100) analyzes the first surveillance image acquired based on the object analysis artificial intelligence model to derive object feature analysis results for objects detected in the first surveillance area (S120). The object analysis artificial intelligence model is trained to recognize objects in the input image and derive object feature analysis results for the recognized objects, including the object category, the object heat map, the object's average movement speed, and the object's location information.

[0127] Next, the wide-angle camera (100) generates a profile for setting the PTZ camera (200) to acquire a second surveillance image for a second surveillance area including the detected object within the first surveillance area in order to obtain a detailed surveillance image for the detected object based on the object feature analysis result derived above (S130). Here, the wide-angle camera (100) may generate a profile to enable the PTZ camera (200) to acquire an image for a certain area (second surveillance area) including the area where the object is detected when a pre-specified object is detected within the first surveillance area as a result of analyzing the first surveillance image. In addition, the wide-angle camera (100) may generate a profile for controlling the PTZ of the PTZ camera (200) further based on the detection frequency of the pre-specified object or the category of the pre-specified object. For example, if the category of the pre-specified object is a vehicle, the wide-angle camera (100) can calculate the moving speed of the pre-specified object and generate a profile for the PTZ settings of the PTZ camera (200) so that the PTZ camera (200) can track the pre-specified object and acquire images according to the calculated moving speed. At this time, if the category of the pre-specified object is a vehicle, the wide-angle camera (100) can generate a profile with the use of a wide-area backlight compensation function added to prevent the vehicle number or passenger identification ability from being reduced due to reflected lighting, backlight, etc.

[0128] The PTZ camera (200) controls the PTZ according to the profile generated by the wide-angle camera (100) and sets the camera to obtain a second surveillance image including the detected object (S140).

[0129] Meanwhile, the PTZ camera (200) analyzes the acquired surveillance image by photographing its surveillance area using a second object analysis artificial intelligence model that has been trained to recognize objects in the input image, and when a pre-designated object is identified within the surveillance area, it can instruct the wide-angle camera (100) to track and store the movement of the object detected in the surveillance area of ​​the wide-angle camera (100). The wide-angle camera (100) can track the object indicated by the wide-angle camera (100) within its surveillance area wider than the angle of view of the PTZ camera (200) and store the movement, and the movement of the stored object can be used for security-related analysis in the future.

[0130] In the above description, the steps, processes, or operations may be further divided into additional steps, processes, or operations, or combined into fewer steps, processes, or operations, depending on the implementation of the present invention. In addition, some steps, processes, or operations may be omitted as needed, or the order between the steps or operations may be switched. In addition, each step or operation included in the automatic camera setting method using the aforementioned artificial intelligence image analysis-based automatic camera setting system may be implemented as a computer program and stored in a computer-readable recording medium, and each step, process, or operation may be executed by a computer device.

[0131] The term "unit" as used herein (e.g., control unit, etc.) can mean a unit that includes one or a combination of two or more of hardware, software, or firmware, for example. "Unit" can be used interchangeably with terms such as unit, logic, logical block, component, or circuit, for example. "Unit" can be the smallest unit of an integrally formed component or a part thereof. "Unit" can also be the smallest unit that performs one or more functions or a part thereof. "Unit" can be implemented mechanically or electronically. For example, "unit" can include at least one of an application-specific integrated circuit (ASIC) chip, field-programmable gate array (FPGA), or programmable-logic device that performs certain operations, which are known or will be developed in the future.

[0132] At least a portion of a device (e.g., modules or functions thereof) or a method (e.g., operations) according to various embodiments may be implemented as instructions stored on a computer-readable storage medium, for example, in the form of a program module. When the instructions are executed by a processor, the one or more processors may perform a function corresponding to the instructions. The computer-readable medium includes all types of recording devices that store data that can be read by a computer system. The computer-readable storage medium / computer-readable recording medium may include a hard disk, a floppy disk, a magnetic media (e.g., a magnetic tape), an optical media (e.g., a compact disc read only memory (CD-ROM), a digital versatile disc (DVD), a magneto-optical media (e.g., a floptical disk), a hardware device (e.g., a read only memory (ROM), a random access memory (RAM), or a flash memory), etc.). In addition, the program instructions may include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc. The above-described hardware devices may be configured to operate as one or more software modules to perform operations of various embodiments, and vice versa.

[0133] Modules or program modules according to various embodiments may include at least one or more of the aforementioned components, some of which may be omitted, or may further include other additional components. Operations performed by modules, program modules, or other components according to various embodiments may be executed sequentially, in parallel, iteratively, or heuristically. Additionally, some operations may be executed in a different order, omitted, or other operations may be added.

[0134] The term "a" as used herein is defined as one or more than one. Furthermore, the use of introductory phrases such as "at least one" and "one or more" in the claims, even if the same claim includes introductory phrases such as "at least one" and "one or more" and the ambiguous phrase "a," should not be construed to mean that the introduction of another claim element by the ambiguous phrase "a" limits any particular claim containing the introduced claim element to an invention containing only one such element.

[0135] In this document, expressions such as "A or B" or "at least one of A and / or B" may include all possible combinations of the items listed together.

[0136] Unless otherwise specified, terms such as "first" and "second" are used arbitrarily to distinguish between the elements they describe. Therefore, these terms are not necessarily intended to indicate temporal or other priority of such elements, nor does the mere fact that certain measures are recited in different claims indicate that a combination of such measures cannot be advantageously employed. Therefore, these terms are not necessarily intended to indicate temporal or other priority of such elements. The mere fact that certain actions are recited in different claims does not indicate that a combination of such actions cannot be advantageously employed.

[0137] The arrangement of components to achieve the same function is effectively "related" to achieve the desired function. Therefore, any two components combined to achieve a specific functionality can be considered "related" to achieve the desired function, regardless of the structural or intermediary components. Similarly, two components thus associated can be considered "operably connected" or "operably coupled" to achieve the desired function.

[0138] Furthermore, those skilled in the art will recognize that the functional boundaries between the aforementioned operations are merely exemplary. Multiple operations may be combined into a single operation, a single operation may be divided into additional operations, and operations may be executed with at least partial temporal overlap. Furthermore, alternative embodiments may include multiple instances of a particular operation, and the order of the operations may be altered in various other embodiments. However, other modifications, variations, and alternatives are also possible. Accordingly, the detailed description and drawings should be considered in an illustrative rather than a restrictive sense.

[0139] The phrase "may be X" indicates that condition X may be satisfied. It also indicates that condition X may not be satisfied. For example, a reference to a system that includes a particular component must also include scenarios where the system does not include the particular component. For example, a reference to a method that includes a particular action must also include scenarios where the method does not include the particular component. However, to take another example, a reference to a system that is configured to perform a particular action must also include scenarios where the system is not configured to perform the particular task.

[0140] The terms "comprising," "having," "consisting of," "consisting of," and "consisting essentially of" are used interchangeably. For example, any method may include at least the acts described in the drawings and / or specification, or may include only the acts described in the drawings and / or specification. Furthermore, the word "comprising" does not exclude the presence of elements or acts listed in a claim.

[0141] Those skilled in the art will recognize that the boundaries between logical blocks are merely exemplary, and that alternative embodiments may merge logical blocks or circuit elements or impose alternative decompositions of functionality across various logical blocks or circuit elements. Therefore, it should be understood that the architecture depicted herein is merely exemplary, and that many other architectures that achieve the same functionality may be implemented.

[0142] Furthermore, for example, in one embodiment, the illustrated examples may be implemented as circuits located on a single integrated circuit or within the same device. Alternatively, the examples may be implemented as any number of individual integrated circuits or individual devices interconnected in any suitable manner, and other variations, modifications, variations, and alternatives are also possible. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense.

[0143] Additionally, for example, the examples described above or portions thereof may be implemented as software or code representations of physical circuits or logical representations convertible to physical circuits, such as in any suitable type of hardware description language.

[0144] Furthermore, the present invention is not limited to physical devices or units implemented with non-programmable hardware, but may also be applied to programmable devices or units capable of performing desired device functions by operating in accordance with appropriate program code, such as mainframes, minicomputers, servers, workstations, personal computers, notepads, personal digital assistants (PDAs), electronic games, automobiles and other embedded systems, mobile phones and various other wireless devices, generally referred to herein as 'computer systems'.

[0145] Connections as described herein may be any type of connection suitable for transmitting signals from or to each node, unit, or device, for example, via an intermediate device. Accordingly, unless explicitly stated otherwise, a connection may be, for example, a direct connection or an indirect connection. A connection may be described or illustrated with reference to a single connection, multiple connections, a unidirectional connection, or a bidirectional connection. However, different embodiments may vary the implementation of the connection. For example, separate unidirectional connections may be used instead of a bidirectional connection, or vice versa. Furthermore, multiple connections may be replaced by a single connection that transmits multiple signals sequentially or in a time-multiplexed manner. Similarly, a single connection transmitting multiple signals may be split into multiple connections that transmit subsets of those signals. Thus, numerous options exist for transmitting signals.

[0146] Preferred embodiments of the technology of this specification have been described above with reference to the attached drawings. The terms and words used in this specification and claims should not be construed as limited to their conventional or dictionary meanings, but rather should be interpreted in their meanings and concepts consistent with the technical spirit of the present invention. The scope of the present invention is not limited to the embodiments disclosed in this specification, and the present invention may be modified, altered, or improved in various forms within the spirit of the present invention and the scope of the claims.

[0147] The present specification can be applied to the fields of surveillance cameras, surveillance camera systems and surveillance service providers, which can automatically set the pan, tilt and zoom of a PTZ camera or automatically set the settings for video capture of a wide-angle camera by using the object recognition results of a wide-angle camera in a camera system composed of a wide-angle camera and a PTZ camera to enhance the surveillance effect of a surveillance area.

Claims

1. A first surveillance camera that uses a wide-angle lens to capture a first surveillance area of ​​a wide area and obtains a first surveillance image; and A second surveillance camera having a PTZ function for capturing a second surveillance area within the first surveillance area to obtain a second surveillance image; The first surveillance camera analyzes the first surveillance image based on the first object analysis artificial intelligence model to derive object feature analysis results for the detected object, and creates a profile for setting the second surveillance camera based on the object feature analysis results. The above second surveillance camera controls PTZ according to the above profile and sets the camera. An automatic camera setting system based on artificial intelligence image analysis.

2. In the first paragraph, the first object analysis artificial intelligence model It is characterized by being trained to recognize an object in an input image and derive the object feature analysis results including the object category, the object heat map, the object average movement speed, and the object location information as the object feature analysis results for the recognized object. An automatic camera setting system based on artificial intelligence image analysis.

3. In the first paragraph, the first surveillance camera, As a result of analyzing the first surveillance image, if a pre-designated object is detected within the first surveillance area, a profile is generated to cause the second surveillance camera to acquire an image of a certain area including the area where the object is detected. An automatic camera setting system based on artificial intelligence image analysis.

4. In the third paragraph, the first surveillance camera, Characterized in that the profile is generated further based on the detection frequency of the above-mentioned pre-specified object or the category of the above-mentioned pre-specified object. An automatic camera setting system based on artificial intelligence image analysis.

5. In paragraph 4, the first surveillance camera, If the category of the above-mentioned pre-designated object is a vehicle, the moving speed of the above-mentioned pre-designated object is calculated, and a profile for the PTZ setting of the second surveillance camera is generated to track the above-mentioned pre-designated object according to the calculated moving speed. An automatic camera setting system based on artificial intelligence image analysis.

6. In paragraph 5, the first surveillance camera, If the category of the above pre-specified object is a vehicle, a profile is created in which the use of the Wide Dynamic Range (WDR) function is further added. An automatic camera setting system based on artificial intelligence image analysis.

7. In the first paragraph, the second surveillance camera, Includes a second object analysis artificial intelligence model trained to recognize objects in input images, The second surveillance video is analyzed using the second object analysis artificial intelligence model, When a pre-designated object is identified within the second surveillance area, the first surveillance camera is instructed to track and store the movement line of the pre-designated object within the first surveillance area. An automatic camera setting system based on artificial intelligence image analysis.

8. In the second paragraph, the first surveillance camera, As a result of analyzing the first surveillance image, a preset profile is recommended based on the category of the object detected within the first surveillance area. An automatic camera setting system based on artificial intelligence image analysis.

9. A camera module that captures a wide surveillance area; and An image analysis module that analyzes a surveillance image of the surveillance area acquired by the camera module based on an object analysis artificial intelligence model, produces an object feature analysis result for a detected object, and generates a profile for setting the camera module based on the object feature analysis result; The above image analysis module sets the camera module according to the above profile. Automatically set up camera based on artificial intelligence image analysis.

10. In the 9th paragraph, the image analysis module, During a predetermined period of time, if an object is recognized in some area within the surveillance area as a result of analyzing the surveillance image, a weight is assigned to each area where the object is recognized based on at least one of the detection frequency of the object and the movement of the object. Generate a profile that sets the image quality acquired differently for each area where the object is recognized based on the above weights, characterized by setting the camera module according to the above profile. Automatically set up camera based on artificial intelligence image analysis.

11. In the 10th paragraph, the image quality is characterized in that it is determined by at least one of the quantization parameters, bitrate and applied codec. Automatically set up camera based on artificial intelligence image analysis.

12. In the 9th paragraph, the image analysis module, During a predetermined period of time, if an object is recognized within the surveillance area as a result of analyzing the surveillance video, the time period in which the object is recognized is determined based on at least one of the detection frequency of the object and the movement of the object. Create a profile that sets the image quality acquired by the camera module differently depending on the time zone, characterized by setting the camera module according to the above profile. Automatically set up camera based on artificial intelligence image analysis.

13. In the 9th paragraph, the image analysis module, As a result of analyzing the above surveillance video, if an object recognized within the surveillance area is identified as a pre-designated object or the movement of the recognized object exceeds a pre-designated standard, a profile is created that sets the image quality of a certain area including the recognized object differently from an area outside the certain area, characterized by setting the camera module according to the above profile. Automatically set up camera based on artificial intelligence image analysis.

14. A camera automatic setting method of an artificial intelligence image analysis-based camera automatic setting system including a first surveillance camera and a second surveillance camera, A step of obtaining a first surveillance image of the first surveillance area by having a first surveillance camera photograph the first surveillance area; A step of analyzing the first surveillance image based on the first object analysis artificial intelligence model by the first surveillance camera to derive object feature analysis results for the detected object; A step for generating a profile for setting the second surveillance camera so that the second surveillance camera obtains a second surveillance image for a second surveillance area within the first surveillance area based on the object feature analysis result of the first surveillance camera; and A method comprising: a step of controlling the PTZ of the second surveillance camera according to the profile and setting the camera to obtain the second surveillance image; 15. In the 14th paragraph, the first object analysis artificial intelligence model A method characterized in that it is learned to recognize an object in an input image and derive an object feature analysis result including an object category, an object heat map, an average movement speed of the object, and an object location information as an object feature analysis result for the recognized object.

16. In paragraph 14, The step of generating a profile for setting the second surveillance camera by the first surveillance camera is as follows: A method comprising: a step of generating a profile so that, when a pre-designated object is detected within the first surveillance area as a result of analyzing the first surveillance image, the second surveillance camera acquires an image of a certain area including the area where the object is detected.

17. In paragraph 16, The step of generating a profile for setting the second surveillance camera by the first surveillance camera is as follows: A method comprising: a step of generating the profile based on the detection frequency of the first surveillance camera or the category of the pre-designated object; 18. In paragraph 17, The step of generating a profile for setting the second surveillance camera by the first surveillance camera is as follows: A method comprising: a step of calculating a moving speed of the pre-designated object if the category of the pre-designated object is a vehicle, and generating a profile for the PTZ settings of the second surveillance camera to track the pre-designated object according to the calculated moving speed.

19. In paragraph 18, The step of generating a profile for setting the second surveillance camera by the first surveillance camera is as follows: A method comprising: a step of generating a profile with added use of a wide-area backlight compensation function if the category of the above-mentioned pre-designated object is a vehicle; 20. In paragraph 14, A step in which the second surveillance camera analyzes the second surveillance video using a second object analysis artificial intelligence model, the second surveillance camera includes a second object analysis artificial intelligence model trained to recognize an object in an input video; and A method further comprising: a step of, when a pre-designated object is identified within the second surveillance area, causing the second surveillance camera to instruct the first surveillance camera to track and store the movement line of the pre-designated object within the first surveillance area.

Images