KR20160037480A - Method for establishing region of interest in intelligent video analytics and video analysis apparatus using the same - Google Patents

Abstract

The present invention relates to a method for establishing a region of interest for intelligent image analysis, capable of automatically establishing a region of interest in accordance with an event rule preset for an input image, and an image analysis apparatus using the method. When an analysis target image is input, a boundary for detecting an event is established on the analysis target image in accordance with an event rule, and a region within a predetermined distance from the boundary is automatically established as a region of interest.

Description

TECHNICAL FIELD [0001] The present invention relates to an intelligent image analyzing apparatus and an image analyzing apparatus,

The present invention relates to intelligent image analysis, and more particularly, to a method and apparatus for setting a region of interest for intelligent image analysis that can automatically set a region of interest according to predetermined event rules for event detection on an input image, ≪ / RTI >

Recently, as the installation of CCTV (Closed Circuit Television) increases, interest for intelligent video analytics is increasing for efficient monitoring. Intelligent image analysis technology is a technology that analyzes the information of images and detects alarms defined in advance and automatically sends alarms to the administrator. Examples of events detected or detected in intelligent image analysis include intrusion detection, access detection, and object count.

The intelligent image analysis includes, for example, background region separation, object detection, object tracking, and event detection. Through this process, predefined events can be detected.

Here, the background region separation is a process of distinguishing a foreground region and a background region of interest from an input image, and object detection is a process of detecting the position, size, and shape of an object in a foreground region, A process of finding a movement path of an object detected in an image, and a process of judging whether an event violates or is satisfied with a predefined rule (hereinafter referred to as an event rule) based on the feature information of the object and the movement information to be.

In this image analysis process, an event rule can be defined as a boundary line set in the image, and an area of the image that performs object detection or tracking according to the event rule is called a region of interest.

However, in the conventional image analysis system, the ROI is arbitrarily set by the administrator, or when the ROI does not specify the ROI, the ROI of the ROI is the ROI. As a result, unnecessary image analysis can be performed. As a result, computer resources are wasted and processing time is increased.

In addition, if the region of interest is not properly set, it may cause an erroneous operation in which an image can not be detected in the image analysis system.

Korean Registered Patent No. 10-1169631, July 24, 2012 (name: image analysis camera with detection interface, image analysis system and detection adjustment method thereof)

Accordingly, the present invention has been proposed in order to efficiently perform intelligent image analysis. More specifically, the present invention relates to an intelligent image analysis method capable of automatically setting an interested region according to predetermined event rules for detecting an event An area setting method and an image analyzing apparatus according to the method.

As a means for solving the above-mentioned problems, the present invention provides a method of analyzing an image, Receiving a boundary setting for detecting an event in the analysis target image; And setting an area within a predetermined distance on the basis of the boundary line as an area of interest for object detection in the analysis target image.

The boundary line may be one of at least one line segment, at least one curve line, a combination of at least one curve line segment, and at least one curve or line segment connected with the line segment, have.

According to another aspect of the present invention, there is provided a method of setting a region of interest for intelligent image analysis, the method further comprising setting the predetermined distance based on the size of the object to adjust the size of the region of interest according to the size of the object .

According to an embodiment of the present invention, the step of setting the region of interest includes: dividing the boundary into one or more segments or curves; Setting a circular region having a predetermined radius as a region of interest as center points of the end points of the at least one line segment or the curve; Setting an area having a predetermined width in the same length as each line segment or curve around the at least one line segment or curve as a region of interest; And establishing a final ROI by connecting the one or more ROIs.

According to an embodiment of the present invention, the step of setting the region of interest includes: generating a distance map indicating a distance between each pixel of the analysis target image and a boundary line based on the boundary line; Extracting pixels having a distance value less than a predetermined value with reference to the distance map; And setting an area including the extracted pixels as a region of interest. In addition, before generating the distance map, a pixel located on the boundary line is rendered in a preset color different from the background image Wherein the step of generating the distance map may calculate a distance value to all pixels of the image up to the closest pixel among pixels having different colors.

In addition, according to another aspect of the present invention, there is provided an image processing apparatus including an event rule setting unit for setting a boundary line for detecting an event in an image to be analyzed; And an interest region setting unit for setting an area within a predetermined distance on the basis of the boundary line as an area of interest for object detection in the analysis target image.

In the image analysis apparatus according to the present invention, the ROI setting unit may set the predetermined distance based on the size of the object, and may adjust the ROI according to the size of the object.

In addition, the ROI setting unit may set an ROI having a predetermined radius as a center of interest by using one or more line segments forming the boundary line or both end points of the ROI as center points, , Sets an area having a predetermined width in the same length as each line segment or a curve as an area of interest, sets a final area of interest by connecting the set one or more areas of interest, A distance map indicating a distance may be generated, and an area including pixels having a distance value less than a predetermined value may be set as a region of interest by referring to the distance map.

The object of the present invention is to set an area of interest, which is an area where object detection and object tracking are to be performed for event detection in the field of intelligent image analysis, and it is an object of the present invention to receive a photographed image as image capturing means such as CCTV, A region within a predetermined distance from the boundary line may be automatically set as a region of interest in the input image.

In addition, according to the present invention, by setting an optimal region of interest based on a boundary line for event detection, it is possible to reduce processing time due to inefficient image analysis, reduce a waste of computing resources, and prevent malfunction of image analysis .

1 is a view schematically showing a configuration of an image surveillance system using intelligent image analysis.
2 is a block diagram illustrating an image analysis apparatus according to the present invention.
3 is a flowchart illustrating an intelligent image analysis process of the image analysis apparatus according to the present invention.
4 is a flowchart illustrating an ROI process for intelligent image analysis according to an exemplary embodiment of the present invention.
5 is a flowchart illustrating a process of setting a region of interest for intelligent image analysis according to another embodiment of the present invention.
FIG. 6 and FIG. 7 illustrate an ROI process according to an embodiment of the present invention.
FIG. 8 is a diagram illustrating an unattractive area removal process in the intelligent image analysis process according to the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the accompanying drawings, detailed description of well-known functions or constructions that may obscure the subject matter of the present invention will be omitted. It should be noted that the same constituent elements are denoted by the same reference numerals as possible throughout the drawings.

The terms and words used in the present specification and claims should not be construed to be limited to ordinary or dictionary meanings and the inventor is not limited to the concept of terminology for describing his or her invention in the best way. It should be interpreted as meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configurations shown in the drawings are merely the most preferred embodiments of the present invention, and not all of the technical ideas of the present invention are described. Therefore, It is to be understood that equivalents and modifications are possible.

Also, terms including ordinal numbers such as first, second, etc. are used to describe various elements, and are used only for the purpose of distinguishing one element from another, Not used. For example, without departing from the scope of the present invention, the second component may be referred to as a first component, and similarly, the first component may also be referred to as a second component.

In addition, when referring to an element as being "connected" or "connected" to another element, it means that it can be connected or connected logically or physically. In other words, it is to be understood that although an element may be directly connected or connected to another element, there may be other elements in between, or indirectly connected or connected.

Also, the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. It is also to be understood that the terms such as " comprising "or" having ", as used herein, are intended to specify the presence of stated features, integers, It should be understood that the foregoing does not preclude the presence or addition of other features, numbers, steps, operations, elements, parts, or combinations thereof.

FIG. 1 shows an example of a video surveillance system using intelligent image analysis technology. The video surveillance system includes an image capturing apparatus 100, an image analyzing apparatus 200, and a monitoring apparatus 300, The image photographed by the photographing apparatus 100 is transmitted to the image analysis apparatus 200 and is analyzed in real time through the image analysis apparatus 200 to detect a predefined event, And outputs it to the device 300. The monitoring apparatus 300 receives an image captured by the image capturing apparatus 100 and displays the captured image in real time and outputs an alarm signal upon occurrence of a previously generated event according to an alarm signal of the image analyzing apparatus 200 do.

As shown in FIG. 1, the present invention is applied to an image analysis apparatus 200 of a video surveillance system for monitoring an image in real time and detecting an event such as entrance or intrusion detection, object count, occurrence of an abnormal situation, To automatically set an area to be subjected to image analysis, that is, a region of interest.

The image analysis apparatus 200 according to the present invention can be configured as shown in FIG.

FIG. 2 is a block diagram of an image analysis apparatus according to an embodiment of the present invention. Referring to FIG. 2, the image analysis apparatus 200 according to the present invention includes an event rule setting unit 210, The interest object extracting unit 240, the event determining unit 250, and the image output unit 260. The image processing unit 260 may include an image processing unit 220, a non-interest area removing unit 230, an interest object extracting unit 240,

Functions and operations of the components constituting the image analysis apparatus 200 according to the present invention will now be described.

The event rule setting unit 210 is a configuration for setting a boundary line for event detection on an image to be analyzed, for example, an image input from the image capturing apparatus 100. [ Here, the boundary line may be one of at least one line segment, at least one curve line, at least one curve line and a combination of line segments, and at least one curve or line segment connected with the boundary line as a reference for applying the event rule . These boundary lines are used for setting a criterion for intrusion detection, access detection, and object count, and may be set according to an input signal of a user (manager) in event rule setting. For example, when the initial event rule is set, the user operates the input means on the basic image input from the image capturing apparatus 100 to select one or more line segments, one or more curves, a combination of one or more curves and line segments, Or a line formed by connecting line segments can be drawn and set. When setting the initial event rule, the event definition information or the detection condition based on the set boundary line may be further set.

In addition, the analysis target image may be a plurality of continuous or discontinuous image frames inputted from the image capturing apparatus 100, or a selected one of image frames captured from the image capturing apparatus 100.

The region-of-interest setting unit 220 is a unit for setting a region of interest in which analysis is performed for detecting or detecting an event among the images to be analyzed. In particular, the ROI setting unit 220 automatically sets a region within a predetermined distance based on the boundary line set by the event rule setting unit 210 as the ROI.

Since the region of interest is an area set for detecting or tracking an object, the region of interest is usually an area sufficiently larger than the object to be detected or tracked. For reference, when expressing the region of interest as a multiple of the width and height of the object, the region of interest should have a size of at least 2 to 5 times the size of the object.

In consideration of this point, a distance value based on the boundary line that affects the size of the ROI must be appropriately set.

However, the size of the object displayed in the image varies depending on the distance between the image and the object, i.e., the object, and thus varies depending on the installation position of the image capturing apparatus 100. Therefore, the size of the object may vary according to the image capturing apparatus 100, and the size of the object appearing in the image varies depending on the capturing position even in the same image capturing apparatus 100. [ In consideration of this point, the present invention can set a distance value that defines the range of the ROI based on the size of the object.

That is, the ROI setting unit 220 may set the distance value based on the size of the currently set object to be detected before setting the ROI.

A method of setting the ROI in the ROI setting unit 220 will be described in more detail with reference to FIGS. 4 and 5. FIG.

Next, the non-interest region removing unit 230 and the remaining region excluding the region of interest of the analysis target image, i.e., the non-interest region requiring no object detection and tracking, are removed. This is a structure for simplifying the processing by omitting the process of determining whether the detected object is included in the ROI in detecting and tracking the object in the ROI extracting unit 240, The pixel values of the remaining non-interest areas are all changed to black so as to block the object detection and tracking in the non-interest area.

To this end, the non-interest area remover 230 generates a mask image for a region of interest. The mask image can be generated according to the algorithm of Equation (1) as an image having a pixel value of 1 in the ROI and a pixel value of 0 in the non-ROI region. In the removal of the non-interest region, the mask image and the image to be analyzed are logically calculated as shown in Equation (2), so that the pixels belonging to the ROI hold the original pixel values and all the pixel values of the ROI are 0 ). ≪ / RTI >

In the above, M (x) denotes a mask image, I (x) denotes an input image, I_M (x) denotes an image from which an unattractive area is removed, and x denotes a pixel.

The interest object extracting unit 240 extracts an interest object of a predefined size from the image in which the non-interest area processed by the non-interest area removing unit 230 is removed, and further, tracks the movement route of the extracted interest object do. In addition, when the non-interest area removing unit 230 is not provided, the interest object extracting unit 240 may further determine whether the extracted object is included in the ROI.

For example, the interest object extracting unit 240 may extract and track a moving part in an image as an object of interest. To do so, the background pixel of the image is designated using a Gaussian model of the pixel , A part not corresponding to the background pixel is divided into the foreground pixels, and the extracted foreground pixel can be used to extract and track the object of interest.

The event determination unit 350 compares the movement state of the object of interest extracted and tracked by the object extraction unit 240 with a preset event rule to determine whether the object of interest has passed, Or the like has occurred.

The image analysis apparatus 200 according to the present invention can transmit the determination result of the event determination unit 350 to the monitoring apparatus 300. [

In addition, the image analysis apparatus 200 according to the present invention can transmit an image designated by the region of interest to the monitoring apparatus 300 through the image output unit 260.

The image output unit 260 transmits the analysis target image combined with the boundary line and the ROI output from the ROI setting unit 220 to the monitoring device 300 and displays the analysis target image on the monitoring device 300, .

Next, an intelligent image analysis process and a ROI setting process performed by the image analysis apparatus 200 configured as described above will be described with reference to FIGS. 3 to 8. FIG.

3 is a flowchart illustrating an intelligent image analysis process of the image analysis apparatus 200 according to the present invention.

Referring to FIG. 3, the image analysis apparatus 200 according to the present invention receives an image to be analyzed (S100). At this time, the analysis target image may be received from the image capturing apparatus 100 shown in FIG. 1 or may be selected from the image transmitted from the image capturing apparatus 100.

When the analysis target image is input, the image analysis device 200 sets a boundary line for detecting an event in the analysis target image according to a user input signal (S110). The boundary line may be set at a designated position of the image to be analyzed, and may be represented by a graphic shape in which one or more line segments, one or more curves, a combination of one or more line segments and curves, and one or more line segments or curves are connected.

In addition, the image analysis apparatus 200 sets the size of the object to be detected or tracked before setting the region of interest (S120), and calculates a distance value d defining the range of the region of interest based on the size of the object (S130). The size of the object may be set according to an input signal of a user (manager), and the distance value d may be set to X times the size (a large value of the width or the length) of the object. Here, X may have a range of 2 to 5, for example.

In step S120, the image analysis apparatus 200 sets an area within a predetermined distance based on one or more line segments or curves included in the boundary line as an area of interest. Here, the predetermined distance based on the one or more line segments or curves corresponds to the distance value set in step S130. The step S120 may be implemented by various methods using a graphic or a distance map, and FIGS. 4 and 5 illustrate the embodiment in detail.

FIG. 4 is a flowchart illustrating a process of setting a region of interest using a graphic. Referring to FIG. 4, in order to automatically set an area of interest, the image analysis apparatus 200 according to the present invention sets boundary lines set in step S110 as one or more line segments or curves (S210).

In step S220, the circular region of interest is created by drawing a circle whose radius is a preset distance value d around each end point of the one or more line segments or curves.

In step S230, a graphic object having a length equal to the length of the one or more line segments or curves and having a height twice the predetermined distance value, that is, 2d, is drawn as an area of interest.

One or more regions of interest may be generated by overlapping with reference to the boundary line through the above-described process.

Accordingly, the image analysis apparatus 200 according to the present invention connects at least one region of interest at least partially overlapped with each other to finally set the region of interest (S240).

6 is a diagram illustrating a process of setting a region of interest using a graphic object according to an embodiment of the present invention. When the boundary line 60 is set as one line segment, as shown in (a), the boundary line 40 Two circular regions of interest 61 and 62 each having a radius d are formed centering on both end points of the boundary line 40. A shape having the same length and a width of 2d around the line segment of the boundary line 40, A region of interest 63 may be generated. The image analysis apparatus 200 according to the present invention concatenates the three regions of interest that are generated as described above and finally generates a region of interest 64 having a shape as shown in FIG. 6 (b).

In addition, the boundary line may be formed by connecting two or more line segments or curves.

7 illustrates a case where the boundary line 70 is formed of two line segments. In this case, as shown in FIG. 7 (a), two line segments are processed in the same manner as in FIG. 6, The region of interest 71, 72 of the shape can be set. Then, the image analysis apparatus 200 according to the present invention connects the two regions of interest 71 and 72 to form a region of interest 73 having a shape as shown in FIG. 7 (b).

Meanwhile, the image analysis apparatus 200 according to the present invention may set a region of interest using a distance map.

5 is a flowchart illustrating a process of setting a region of interest using a distance map according to another embodiment of the present invention.

Referring to FIG. 5, the image analysis apparatus 200 according to the present invention renders a pixel corresponding to a boundary line defined in an event rule to be displayed in a specific color (S310). The specific color may be a color not appearing in a complementary color or background image of the background image of the analysis target image. For example, if the background image is bright, the pixel corresponding to the boundary line is set to black, and if the background image is dark, the pixel corresponding to the boundary line can be set to white.

Thereafter, the image analysis apparatus 200 generates a distance map based on the boundary line (S320). More specifically, the image analysis apparatus 200 calculates the distance between each pixel of the image to be analyzed and the pixel having the closest specific color. Thus, the distance to the boundary line can be obtained for each pixel of the image to be analyzed.

Then, the image analysis apparatus 200 sets an area including all the pixels whose distance value calculated from the distance map is equal to or less than d as an area of interest (S330).

4 or FIG. 5, the image analysis apparatus 200 according to the present invention can set a region of interest that includes a boundary line and is sufficiently larger than an object.

After the region of interest is set as described above, the image analysis apparatus 200 according to the present invention should extract an object of interest for detecting an event from a region of interest.

To this end, the image analysis apparatus 200 according to the present invention may remove the remaining region excluding the region of interest, that is, the non-interest region, from among the analysis target images as shown in FIG. 3 (S150). 8 (a), the pixels included in the ROI have pixel values of 1 (white), and the remaining pixels have 0 (black The mask image 81 having the pixel value of " 0 " The pixel values of the mask image 81 and the analysis target image are logically calculated to maintain the pixel values of the ROIs in the analysis target image and change the pixel values of the remaining regions to zero. According to this, the image 82 in which the non-interest area is removed as shown in FIG. 8B is created.

After the non-interest area is removed from the analysis target image, the image analysis apparatus 200 according to the present invention extracts the interest object from the analysis target image from which the non-interest area is removed as shown in FIG. 3 , And the extracted interest object is tracked (S160). In this case, since the image analyzing apparatus 200 can extract the object of interest by processing only the region of interest of the analysis target image, unnecessary waste of computing resources can be reduced and the processing time can be shortened.

Thereafter, the image analysis apparatus 200 may compare the extracted interest object and the tracking path with preset event rules to determine whether an event has occurred (S170).

The method of setting a region of interest according to the present invention may be implemented in software readable by various computer means and recorded in a computer-readable recording medium. Here, the recording medium may include program commands, data files, data structures, and the like, alone or in combination. Program instructions to be recorded on a recording medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. For example, the recording medium may be an optical recording medium such as a magnetic medium such as a hard disk, a floppy disk and a magnetic tape, a compact disk read only memory (CD-ROM), a digital video disk (DVD) Includes a hardware device that is specially configured to store and execute program instructions such as a magneto-optical medium such as a floppy disk and a ROM, a random access memory (RAM), a flash memory, do. Examples of program instructions may include machine language code such as those generated by a compiler, as well as high-level language code that may be executed by a computer using an interpreter or the like. Such a hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be apparent to those skilled in the art. Furthermore, although specific terms are used in this specification and the drawings, they are used in a generic sense only to facilitate the description of the invention and to facilitate understanding of the invention, and are not intended to limit the scope of the invention.

In addition, the image analysis apparatus 200 according to the present invention can process a program command for executing the method according to the present invention through a processor to perform the functions according to the present invention. In one implementation, the processor may be a single-threaded processor, and in other embodiments, the processor may be a multithreaded processor. Further, the processor is capable of processing instructions stored on a memory or storage device.

Although the present specification and drawings describe exemplary device configurations, the functional operations and subject matter implementations described herein may be embodied in other types of digital electronic circuitry, or alternatively, of the structures disclosed herein and their structural equivalents May be embodied in computer software, firmware, or hardware, including, or in combination with, one or more of the foregoing. Implementations of the subject matter described herein may be embodied in one or more computer program products, i. E. One for computer program instructions encoded on a program storage medium of the type for < RTI ID = 0.0 & And can be implemented as a module as described above. The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter that affects the machine readable propagation type signal, or a combination of one or more of the foregoing.

While the specification contains a number of specific implementation details, it should be understood that they are not to be construed as limitations on the scope of any invention or claim, but rather on the description of features that may be specific to a particular embodiment of a particular invention Should be understood. Certain features described herein in the context of separate embodiments may be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented in multiple embodiments, either individually or in any suitable subcombination. Further, although the features may operate in a particular combination and may be initially described as so claimed, one or more features from the claimed combination may in some cases be excluded from the combination, Or a variant of a subcombination.

Likewise, although the operations are depicted in the drawings in a particular order, it should be understood that such operations must be performed in that particular order or sequential order shown to achieve the desired result, or that all illustrated operations should be performed. In certain cases, multitasking and parallel processing may be advantageous. Also, the separation of the various system components of the above-described embodiments should not be understood as requiring such separation in all embodiments, and the described program components and systems will generally be integrated together into a single software product or packaged into multiple software products It should be understood.

Certain embodiments of the subject matter described herein have been described. Other embodiments are within the scope of the following claims. For example, the operations recited in the claims may be performed in a different order and still achieve desirable results. By way of example, the process illustrated in the accompanying drawings does not necessarily require that particular illustrated or sequential order to obtain the desired results. In certain implementations, multitasking and parallel processing may be advantageous.

The description sets forth the best mode of the invention, and is provided to illustrate the invention and to enable those skilled in the art to make and use the invention. The written description is not intended to limit the invention to the specific terminology presented. Thus, while the present invention has been described in detail with reference to the above examples, those skilled in the art will be able to make adaptations, modifications, and variations on these examples without departing from the scope of the present invention.

Therefore, the scope of the present invention should not be limited by the described embodiments but should be defined by the claims.

The object of the present invention is to set an area of interest as an area to be analyzed for event detection such as object detection and object tracking in the field of intelligent image analysis. A boundary line for detecting an event is set in the image and an area within a predetermined distance based on the boundary line in the input image can be automatically set as a region of interest.

In addition, according to the present invention, by setting an optimal ROI based on a boundary line defined in an event rule, it is possible to reduce processing time due to inefficient image analysis and waste of computing resources.

100: image capturing apparatus 200: image analyzing apparatus
300: Monitoring device 210: Event rule setting unit
220: ROI setting unit 230: Non-ROI removing unit
240: Interest object extraction unit 250: Event processing unit
260: Video output section

Claims (10)

Receiving an analysis target image;
Receiving a boundary setting for detecting an event in the analysis target image;
And setting an area within a predetermined distance on the basis of the boundary line as an area of interest for object detection in the analysis target image. The method according to claim 1,
Wherein the boundary line is one of at least one line segment, at least one curve, a combination of at least one curve and line segment, and at least one curve or line segment connected to the boundary line. The method according to claim 1,
Further comprising setting the predetermined distance based on the size of the object, wherein the size of the region of interest is adjusted according to the size of the object. 2. The method of claim 1, wherein setting the region of interest comprises:
Dividing the boundary line into one or more line segments or curves;
Setting a circular region having a predetermined radius as a region of interest as center points of the end points of the at least one line segment or the curve;
Setting an area having a predetermined length and a length equal to the length of each line segment or curve around the at least one line segment or curve as a region of interest;
And establishing a final region of interest by connecting the at least one region of interest to the region of interest. 2. The method of claim 1, wherein setting the region of interest comprises:
Generating a distance map indicating a distance between each pixel of the image to be analyzed and a boundary line;
Extracting pixels having a distance value less than a predetermined value with reference to the distance map;
And setting an area including the extracted pixels as a region of interest. 6. The method of claim 5, wherein setting the region of interest comprises:
Further comprising rendering the pixel located on the boundary line in a predetermined color before generating the distance map,
Wherein the step of generating the distance map calculates a distance value to each pixel of the analysis target image to a closest pixel among the pixels having the hue. An event rule setting unit for setting a boundary line for detecting an event in the analysis target image; And
And an interest region setting unit that sets an area within a predetermined distance from the analysis target image as an area of interest for object detection based on the boundary line. 8. The apparatus according to claim 7, wherein the ROI setting unit
Wherein the predetermined distance is set based on the size of the object so that the region of interest is adjusted according to the size of the object. 8. The apparatus according to claim 7, wherein the ROI setting unit
A circular region having a predetermined radius is set as a region of interest, with both end points of the line segment or the curve forming the boundary as a center point, and the length of each line segment or curve around the one or more line segments or curves Sets an area having the same length and a predetermined width as a region of interest, and connects the set at least one region of interest to set a final region of interest. 8. The apparatus according to claim 7, wherein the ROI setting unit
A distance map indicating a distance between each pixel of the image to be analyzed and a boundary line is generated and an area including pixels having a distance value less than a predetermined value is set as an area of interest with reference to the distance map, Image analysis device.

Images