JP4859795B2 - Camera device and recording system - Google Patents

Description

  The present invention relates to a camera device that is installed in, for example, an elevator car and photographs the inside of the car.

  As a conventional technology of a security camera installed in an elevator car, there is one that installs a security camera on a car and takes an image of the inside of the car reflected by a reflector (for example, see Patent Document 1). Thus, even a security camera with a narrow angle of view can shoot a wide range in the car. However, even with this configuration, it is not possible to shoot the entire car. Although it is possible to take a wide range of images by using a reflecting mirror as a convex mirror or the like, the object to be imaged becomes smaller and the identification becomes difficult.

When a general security camera is used, the security camera is a display device such as a CRT (Cathode ray tube) or LCD (Liquid Crystal Display) or a recording device such as a VTR (Video Tape Recorder) or HDD (Hard Disk Drive) recorder. Connected to. In general, the image output uses industry standard analog signals such as NTSC (National TV Standards Committee) and PAL (Phase Alternating Line). Analog signals such as NTSC and PAL are not so high in resolution (in the case of NTSC, about 450 vertical resolutions and about 320 horizontal resolutions), and when a wide range of images / videos are recorded, there is a problem that it becomes difficult to identify details. . There are also products in which the output signal to the display device is converted into a digital signal, but in that case, the format of the digital signal is different for each manufacturer.
Japanese Utility Model Publication No. 5-95967

Since the security camera of the elevator is close to the object to be photographed, a blind spot is formed in the car at the angle of view of a normal security camera. If you shoot with a security camera equipped with a wide-angle lens to eliminate blind spots and fit the whole image into image standards such as NTSC and PAL, the amount of information on feature points such as “moving points” and “human faces” is small. Therefore, there is a problem that it is difficult to recognize feature points as an image. On the other hand, if the amount of information is such that the feature points can be recognized, there is a problem that the information cannot be accommodated in a transmission method such as NTSC or PAL.
There is a method of shooting the entire interior of the car by panning / tilting a camera with a narrow angle lens with a motor or the like.

  The present invention has been made in order to solve, for example, the above-described problem, and an object of the present invention is to provide a camera device that can output a detailed image without missing an event necessary for crime prevention. To do.

The camera device according to the present invention includes:
A photographing unit that photographs a predetermined range and inputs an image of the photographed predetermined range as a wide-range image;
An image processing unit for processing the wide range image input by the imaging unit;
An image output unit that outputs at least a part of the wide range image processed by the image processing unit as an output image;
The image processing unit
Feature points indicating features of the wide range image are detected from the wide range image input by the photographing unit by a predetermined detection procedure, and feature point type information indicating the type of the feature point based on the detection result and the wide range image A feature point detection unit that determines the feature point position information indicating the position of the feature point in and outputs the determined feature point type information and the feature point position information;
An image output range determination unit that inputs the feature point type information and the feature point position information, and determines an output range by a predetermined range determination procedure using the feature point type information and the feature point position information;
An image extraction unit that extracts an image included in the output range determined by the image output range determination unit as an extracted image from the wide range image input by the imaging unit, and outputs the extracted image to the image output unit; Prepared,
The image output unit includes:
The extracted image extracted by the image extraction unit is output as the output image.

  According to the camera device of the present invention, the image processing unit detects feature points indicating image movement, a human face, and the like from a wide range image captured by the imaging unit, and determines an output range (image output range). The extracted image is extracted from the wide range image and the image output range, and is output to the image output unit. Therefore, it is possible to transmit and display with a quantity of information that can identify feature points on a display device such as NTSC or PAL having a low resolution. In addition, it is possible to display feature points in more detail by detecting feature points from a wide-range image and outputting a narrow range centering around the feature points as an output image.

Embodiment 1 FIG.
Hereinafter, embodiments of the present invention will be described with reference to FIGS.
FIG. 1 is a block configuration diagram showing a configuration of a recording system including a camera device according to an embodiment described below. FIG. 2 is a diagram illustrating an example of a hardware configuration of the camera device and the recording system according to the first embodiment. FIG. 3 is a flowchart showing the operation of the camera device according to the first embodiment.

  The recording system 1000 includes a camera device 500, a display device 400 such as a CRT or LCD, and a recording device 920 such as a VTR or HDD recorder. The camera device 500 is connected to the display device 400 or the recording device 920. As a general image output signal output from the camera device 500 to the display device 400 or the recording device 920, an industry standard analog signal such as NTSC or PAL is used. There is also a product in which the image output signal from the camera device 500 is converted into a digital signal. When both the recording device 920 and the display device 400 support the digitized signal, the camera device 500 outputs the digitized image output signal. It doesn't matter.

  A camera device 500 according to the present embodiment will be described. The camera device 500 includes a photographing unit 200, an image identification circuit 100 (an example of an image processing unit), a timer 180, and an image output unit 300.

  The imaging unit 200 includes a lens, a CCD (Charge-Coupled Devices), a CMOS (Complementary Metal Oxide Semiconductor) image sensor, and the like, and converts an optical image into an electrical signal. When the lens has a wide angle, it is possible to capture a wide range. A predetermined range is photographed and an image of the photographed predetermined range is input as a wide range image. When the camera device 500 is installed in an elevator car, the entire area of the elevator car can be photographed with a wide-angle lens and input as a wide-range image.

The image identification circuit 100 processes a wide range image captured and input by the imaging unit 200.
The timer 180 is a device that measures and measures time. The timer 180 detects an imaging cycle that is an interval at which the imaging unit 200 captures a predetermined range, and an interval at which the image input circuit 100 processes an image (for example, a feature point is detected. Measurement interval) is measured.

  The image output unit 300 inputs an image processed by the image identification circuit 100 and outputs it as an output image. The image output unit 300 outputs the output image in accordance with the method of the display device 400 or the recording device 920 as the output destination. For example, when the display device 400 is a television, it is converted into an analog signal such as NTSC or PAL and output. When the display device 400 or the recording device 920 connected to the camera device 500 is a digital signal system, the image output unit 300 outputs a digital signal to the display device 400 or the recording device 920.

  The image identification circuit 100 includes an image extraction unit 110, an image output range determination unit 120, and a feature point detection unit 130. An image signal (an example of a wide range image) captured by the imaging unit 200 is sent (inputted) to the feature point detection unit 130 and the image extraction unit 110.

The feature point detection unit 130 detects feature points indicating features from a wide range image by a predetermined detection procedure. In addition, the feature point detection unit 130 detects feature points for each predetermined detection period measured by the timer 180.
The predetermined detection procedure is, for example, a procedure for detecting a moving object as a feature point from an image periodically input by image analysis, or detecting a human face or hand as a feature point by image pattern matching or the like. There are detection procedures such as image processing, detection of opening / closing of elevator doors and changes in indicators.
When the feature point detection unit 130 detects a feature point, the feature point detection unit 130 further determines what the feature point is based on the detection result, and feature point type information indicating the determination result Is generated. The feature point type information is, for example, information “human face, hand, luggage, unknown”. The feature point detection unit 130 determines (determines) whether the feature point is a human face or a carried baggage. At the same time, the feature point detection unit 130 determines (calculates) feature point position information indicating the position of the feature point in the wide-range image. The detected feature points generally have a certain range. As the feature point range information, for example, the feature point range information may be the coordinates of the points located at the top, bottom, left, and right of the feature point range detected as a cluster. The feature point detection unit 130 calculates the center (center of gravity) from the detected feature point range information and determines the center point. Information including the coordinates of the center point in the wide-range image and the information on the range of the feature points is used as the feature point position information.
After detecting the feature points, the feature point detection unit 130 sends the feature point position information and the feature point type information to the image output range determination unit 120.

The image output range determining unit 120 extracts and enlarges at least a part of an image from a wide range image from the feature point position information and feature point type information of the feature points (hereinafter referred to as an output range or an image output range). ) And the image output range is sent to the image extraction unit 110 for each detection cycle.
The image extraction unit 110 uses the image output range sent from the image output range determination unit 120 to extract a part of an image of the entire shooting range (an example of a wide range image) sent from the shooting unit 200 for each shooting period. To the image output unit 300.

  The image output unit 300 converts the extracted image extracted by the image extraction unit 110 into an analog signal such as NTSC or PAL as an output image, and outputs the analog signal.

Here, the imaging cycle and the detection cycle measured by the timer 180 will be described. The imaging unit 200 repeatedly captures a predetermined range in accordance with the timing of the imaging cycle measured by the timer 180 and outputs it to the image identification circuit 100 as a wide range image. Further, the image identification circuit 100 processes the input wide-range image in accordance with the detection cycle timing measured by the timer 180. The imaging cycle and the detection cycle may be the same cycle. However, the time required for the image identification circuit 100 to process the wide range image may take a long time compared to the time for the imaging unit 200 to capture the wide range image. When the processing time of the image identification circuit 100 is long, for example, the detection cycle is a multiple of the shooting cycle (when the shooting cycle is 1/30 second, the detection cycle is set to a value such as 1/5 second). . By using a multiple, it is possible to synchronize the shooting timing and the processing timing at regular intervals.
For example, the image identification circuit 100 inputs a wide range image in one detection cycle, performs a process until the feature point detection unit 130 detects the feature point, and the image output range determination unit determines the image output range. . The image extraction unit 110 extracts an extracted image from the wide-range image and the image output range in accordance with the shooting cycle, and outputs the extracted image to the image output unit 300. Since the image extraction unit 110 extracts the extracted image in accordance with the shooting cycle and outputs the extracted image to the image output unit 300, if the detection cycle is longer than the shooting cycle, the image output range determination unit 120 sets the new image output range. The image extraction unit 110 extracts the extracted image from the wide range image using the same image output range.

Next, an example of hardware resources of the recording system 1000 will be described with reference to FIG.
In FIG. 2, the recording system 1000 includes a CPU 911 (also referred to as a central processing unit, a central processing unit, a processing unit, or an arithmetic unit) that executes a program. The CPU 911 is connected to the ROM 913, the RAM 914, the communication board 915, the display device 400, the input device 902, the recording device 920, and the camera (the photographing unit 200) via the bus 912, and controls these hardware devices. The recording device 920 is, for example, a magnetic disk device, an optical disk device, a memory card read / write device, an HHD recorder, or a VTR.
The RAM 914 is an example of a volatile memory. The storage medium of the ROM 913 and the recording device 920 is an example of a nonvolatile memory.
The communication board 915, the input device 902, the camera (the photographing unit 200), and the like are examples of the input unit and the input device.
The communication board 915, the display device 400, and the like are examples of an output unit and an output device.

The communication board 915 is connected to the facsimile machine 932, the telephone 931, the LAN 942, and the like. The communication board 915 is not limited to the LAN 942 and may be connected to the Internet 940, a WAN (wide area network) such as ISDN, or the like.
The recording device 920 stores an operating system 921 (OS), a program group 922, and a file group 923. The programs in the program group 922 are executed by the CPU 911 and the operating system 921.

The program group 922 stores a program for executing a function described as “˜unit” in the description of the embodiment. The program is read and executed by the CPU 911.
In the file group 923, data, signal values, variable values, and parameters described as “determination result of”, “calculation result of”, and “processing result of” in the description of the embodiment are “to file”. And “˜database”.
In addition, the arrows in the flowcharts described in the description of each embodiment mainly indicate input / output of data and signals. The data and signal values are the memory of the RAM 914, the magnetic disk of the recording device 920, other optical disks, mini disks, It is recorded on a recording medium such as a DVD (Digital Versatile Disk). Data and signals are transmitted online via a bus 912, signal lines, cables, or other transmission media.

  In addition, what is described as “unit” in the description of the embodiment described below may be realized by firmware stored in the ROM 913. Alternatively, it may be implemented by software alone, hardware alone, a combination of software and hardware, or a combination of firmware. Firmware and software are stored as programs in a recording medium such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, and a DVD. The program is read by the CPU 911 and executed by the CPU 911. That is, the program causes the computer to function as “to part” described below. Alternatively, it causes the computer to execute the procedure of “to part” described below.

  Next, an example of the operation of the camera device 500 according to the present embodiment will be described with reference to FIG.

  First, in step S101 (wide range image capturing process), the capturing unit 200 of the camera device 500 captures a wide range image. The wide-range image is image information obtained by photographing the entire range in the elevator car through, for example, a wide-angle lens.

Next, in step S102 (feature point detection processing), the feature point detection unit 130 inputs a wide range image from the photographing unit 200, and performs image analysis using the CPU. When the feature point detection unit 130 detects feature points of a wide range image by image analysis (in the case of Yes), the feature point detection unit 130 determines feature point type information and feature point position information.

First, the feature point detection unit 130 determines what kind of feature point is based on the result of image analysis, and determines feature point type information indicating the type of the feature point based on the determined result. The type of feature point is information that distinguishes whether the feature point is a person's face or a baggage with movement that is not a person, for example. The type of feature point is determined by the feature point detection unit 130 performing image analysis on a wide range image. An example of image analysis is pattern matching. Pattern matching is, for example, storing an image of feature points as a specific pattern in advance, searching whether there is an image similar to the specific pattern in the wide-range image, and approximately matching the specific pattern This is a method for discriminating that a feature point meaning a specific pattern is included when it is found.
Also, the feature point detection unit 130 determines feature point position information. The feature point position information is, for example, information indicating a certain range of feature points and the coordinates of the center point of the feature points. A feature point is detected as a block of pixels (pixels) having a certain range from a wide range image. As information indicating the range of feature points, for example, there is a method of representing the coordinates of the pixels at the top, bottom, left, and right ends among pixels detected as feature points. The feature point detection unit 130 calculates the center point of the feature point from the feature point range information and includes it in the feature point position information. For the center point, the Y coordinate of the middle point of the uppermost pixel may be the Y coordinate of the center point, and the X coordinate of the middle point of the leftmost pixel may be the X coordinate of the center point.
The feature point detection unit 130 that has detected the feature point inputs the determined feature point position information and feature point type information to the image output range determination unit 120.

  When the feature point detection unit 130 does not detect the feature point in step S102 (in the case of No), the process moves to step S106. In step S <b> 106, the image extraction unit 110 outputs the wide range image to the image output unit 300 as an extracted image. In step S106, the wide-range image having no feature points means that there is no abnormality in the elevator car. Therefore, for example, the image extraction unit 110 may input a signal indicating no abnormality to the image output unit instead of outputting the extracted image. In this case, nothing is recorded in the recording device 920, or only a signal indicating no abnormality is recorded. In this case, the storage capacity of the recording device 920 is saved.

In step S103 (image output range determination processing), the image output range determination unit 120 outputs an image based on feature point information (feature point position information and feature point type information) input from the feature point detection unit 130. Determine the range. The image output range is information for the image extraction unit 110 to cut out a part of the wide range image input from the photographing unit 200 in the next step 104. The shape of the image output range depends on the recording method recorded by the recording device 920 and the display method of the display device 400. For example, in the case of an NTSC signal, the aspect ratio is 4: 3. As a method for determining the image output range, the size is determined based on the feature point type information with the input feature point position information as a center. For example, if the feature point is a human face, the image output range is determined so that the maximum feature point is included from the feature point range information included in the feature point position information so that the human face can be identified. And If it is unclear what the feature points are, the image output range is set to a range wider than the feature point range information so that the surroundings can be seen to some extent.
After determining the image output range, the image output range determination unit 120 outputs the image output range to the image extraction unit 110. The image extraction unit 110 inputs an image output range from the image output range determination unit 120.

  In step S <b> 104 (image extraction processing), the image extraction unit 110 uses the image output range input from the image output range determination unit 120 from the wide range image input from the imaging unit 200, and includes images included in the image output range. To extract. The image extracted by the image extraction unit 110 is called an extracted image. By extracting an image in the image output range from a wide-range image, it is possible to output an image with enlarged feature points. The image extraction unit 110 outputs the extracted extracted image to the image output unit 300. output.

  In step S105 (image output processing), the image output unit 300 outputs the image (extracted image) input by the image extraction unit 110 as an output image to the recording device 920 or the display device 400. If no feature point is detected in step S102, the wide range image captured by the imaging unit 200 is output as an output image.

The camera device 500 described in this embodiment includes:
An image capturing unit 200 that captures a predetermined range and inputs the captured image of the predetermined range as a wide range image, an image processing unit (image identification circuit 100) that processes the wide range image input by the image capturing unit 200, and an image processing unit An image output unit 300 that outputs at least a part of the wide range image processed by the (image identification circuit 100) as an output image;
The image processing unit (image identification circuit 100) detects a feature point indicating the feature of the wide-range image from the wide-range image input by the photographing unit 200 using a predetermined detection procedure, and indicates the type of the feature point based on the detection result. A feature point detection unit 130 that determines the feature point type information and the feature point position information indicating the position of the feature point in the wide-range image, and outputs the determined feature point type information and the feature point position information;
An image output range determination unit 120 that inputs feature point type information and feature point position information, determines an output range by a predetermined range determination procedure using the feature point type information and the feature point position information,
An image extraction unit 110 that extracts an image included in the output range determined by the image output range determination unit 120 as an extracted image from the wide range image input by the photographing unit 200 and outputs the extracted image to the image output unit. ,
The image output unit 300 outputs the extracted image extracted by the image extracting unit 110 as the output image.

The camera device 500 further includes
A timer 180 for measuring an imaging cycle indicating an interval at which the imaging unit 200 captures an image and a detection cycle indicating an interval at which the feature point detection unit 130 detects a feature point;
The imaging unit 200 captures a predetermined range for each imaging cycle measured by the timer 180, inputs an image of the captured predetermined range as a wide range image,
The feature point detection unit 130 detects a feature point indicating a feature of the wide-range image input from the imaging unit 200 for each detection cycle measured by the timer 180 by a predetermined detection procedure, and the feature point based on the detection result. Feature point type information indicating the type of the feature point and feature point position information indicating the position of the feature point in the wide-range image,
The image output range determination unit 120 inputs feature point type information and feature point position information for each detection period, and determines an output range by a predetermined range determination procedure using the feature point type information and feature point position information. And
The image extraction unit 110 extracts, as an extracted image, image information included in the output range determined by the image output range determination unit from the wide range image input by the photographing unit 200 for each detection cycle, and extracts the extracted image information to the image output unit 300. An image is output.

According to the camera device 500 described in the present embodiment, the image processing unit (image identification circuit 100) identifies feature points indicating image movement, a human face, and the like from a wide range image captured by the imaging unit 200. The image output range is determined, an image within the image output range is extracted from the wide range image, and is output to the image output unit 300 as an output image. Therefore, it is possible to transmit and display with a quantity of information that can identify feature points on a display device such as NTSC or PAL having a low resolution. In addition, by detecting feature points from a wide-range image and outputting a narrow range centering around the feature points as an output image, the display device 400 can display the feature points in more detail.
In addition, by providing the timer, the photographing unit shoots at a predetermined shooting cycle, so that the feature points can be recorded as a moving image. In addition, since the process for detecting the feature points may take time, the shooting cycle and detection are performed by separately providing the shooting cycle for shooting by the shooting unit and the detection cycle for detecting the feature points by the feature point detection unit 130. It is possible to set with a period different from the period. For example, the imaging cycle is 1/30 second, but the detection cycle can be 1/5 second.

Embodiment 2. FIG.
In the first embodiment, the image extraction unit 110 extracts an extracted image from a wide range image captured by the imaging unit 200 with a wide-angle lens and an image output range centered on a feature point, and the image output unit 300 outputs the extracted image. The image is output to the recording device 920 or the display device 400 as an image. The camera device according to the present embodiment is characterized in that the camera device 500 described in the first embodiment further includes an image correction unit that corrects distortion of an image captured by a wide-angle lens.
Hereinafter, the present embodiment will be described with reference to FIGS.

  FIG. 4 is a block configuration diagram showing the configuration of the camera device 500 according to the second embodiment. FIG. 5 is a flowchart showing the operation of the camera device 500 according to the second embodiment. In the following description, differences between the present embodiment and the first embodiment will be mainly described. In FIG. 4, the same reference numerals are given to functional blocks showing the same functions and operations as those in the first embodiment, and the description thereof is omitted. In FIG. 4, the difference from the first embodiment is that the image identification circuit 100 of the camera device 500 includes an image distortion correction unit 140.

  The imaging unit 200 captures and inputs a wide range image through a wide-angle lens. The imaging unit 200 may capture a wide range image using a mirror (for example, a curved mirror). As the shooting range of the shooting unit 200 becomes wider and approaches 180 degrees, distortion occurs in the surrounding images (images far from the center of the shooting range, peripheral portions of the wide range image). This always occurs in order to apply a three-dimensional space to a two-dimensional screen. When the image extraction unit 110 extracts an extracted image from a wide-range image, an image with almost no distortion can be obtained if the range in which the image is enlarged is narrowed. However, the surrounding image captured at a wide angle is already distorted, and even if it is simply enlarged, the image remains distorted. In view of this, when the extracted image is enlarged, it is not set to the same magnification, but the image distortion is corrected to make the image easy to see.

  The image distortion correction unit 140 inputs the extracted image extracted by the image extraction unit 110. The image distortion correction unit 140 corrects the distortion of the extracted image and outputs the corrected image to the image output unit 300 as a corrected image.

The correction coefficient (parameter) used when correcting the distortion differs depending on the position extracted from the wide range image photographed with the wide-angle lens. The central part of the wide-range image has a small distortion, but the edge part has a large distortion.
Therefore, when correcting the distortion of the extracted image, the image distortion correction unit 140 may acquire the image output range determined by the image output range determination unit and obtain the correction coefficient from the image output range.

  Next, the operation of the camera device 500 will be described with reference to FIG. The same processing contents as those of the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.

In FIG. 5, the difference from the first embodiment is that step S107 (distortion correction processing) is added.
In step S104, the image extraction unit 110 extracts an extracted image from the wide range image based on the image output range. The extracted extracted image is input from the image extraction unit 110 to the image distortion correction unit 140.
In step S107 (distortion correction processing), the image distortion correction unit 140 receives the extracted image and creates a corrected image with distortion corrected. The image distortion correction unit 140 inputs the corrected image to the image output unit 300.
In step S105, the image output unit 300 outputs the input corrected image as an output image to the recording device 920 or the display device 400.

The camera device 500 according to the present embodiment is
The imaging unit 200 captures a predetermined range (for example, in an elevator car) at a wide angle, and the image identification circuit 100 (image processing unit) further inputs the extracted image from the image extraction unit 110, and the imaging unit 200 captures the wide-angle lens. The image output unit 300 includes an image distortion correction unit 140 that corrects distortion of an extracted image generated by photographing a predetermined range using a wide angle and outputs the corrected image to the image output unit 300 as a corrected image. The correction image is input from the image distortion correction unit 140 and output as an output image.

  The camera apparatus 500 according to the present embodiment includes the image distortion correction unit 140 that corrects distortion of an image photographed at a wide angle, and thus can output an image with almost no distortion. Therefore, even when the camera device 500 is provided in an elevator car for crime prevention purposes, an easy-to-view image without distortion is output to a display device or the like. That is, there is an effect that the state in the car can be clearly confirmed.

Embodiment 3 FIG.
In the first and second embodiments, the feature point detection unit 130 detects a feature point for each detection cycle measured by the timer 180, and the image output range determination unit 120 uses the feature point position information and feature point type information of the feature point. The camera apparatus 500 that determines the image output range has been described. In this embodiment, for example, feature point position information of feature points for each detection cycle is stored, the amount of feature point movement is calculated as the amount of feature point movement, and the image output range is determined based on the amount of feature point movement. A camera device 500 that is characterized in that will be described.

  The output image output by the camera device 500 described in the first embodiment is an image extracted from the wide range image captured by the imaging unit 200 with the feature point as the center. Therefore, when the feature point moves greatly instantaneously, the image output range determined by the image output range determination unit 120 also moves instantaneously, and the image captured in the output image finally output from the image output unit 300 also Move instantly. Although the feature points are photographed in an enlarged manner, there is a problem that it is difficult to determine from the output image where the feature points are because the surrounding situation changes instantaneously.

  Therefore, a description will be given below of a camera device 500 that is intended to output an image that is easy to see without instantaneously switching the image output by the image output unit 300 even when the feature point moves instantaneously.

  Since the configuration and hardware configuration of the camera device 500 according to the present embodiment are the same as those of the first embodiment, description thereof is omitted.

The difference from the camera device 500 shown in the first embodiment is that feature point position information of feature points detected by the feature point detection unit 130 for each detection cycle is stored, and an image output range determination unit for each detection cycle. The image output range is stored.
The feature point detection unit 130 detects a feature point for each detection period measured by the timer 180, and further stores feature point position information indicating the position information of the feature point. Then, for each detection cycle, the stored feature point position information of the previous cycle is compared with the feature point position information of the feature point detected at the present time, and the movement amount is calculated as the feature point movement amount. The feature point movement amount calculated by the feature point detection unit 130 is input to the image output range determination unit 120.

  The image output range determination unit 120 determines the image output range based on the feature point position information and the feature point type information, as well as the feature point movement amount calculated and input by the feature point detection unit 130. When the amount of movement of the feature point is small, the details of the feature point are grasped. Therefore, the width of the feature point is calculated using the X coordinate of the point located on the left and right of the feature point included in the feature point position information. The image output range is determined so as to include the maximum number of points. For example, the width of the feature point and the width of the image output range are made substantially the same. When the feature point moving amount is large, the image output range is set wider than the feature point range. By setting widely, the feature point and the situation around the feature point are included in the output image, and the situation around the feature point can be easily grasped. For example, the length (distance) twice the width of the feature point is set as the width of the image output range. The image output range may be obtained based on the height of the feature point (obtained from the Y coordinate of the uppermost and lowermost feature points). The image output range may be determined based on the larger of the feature point width and height.

Further, the image output range determination unit 120 stores an image output range for each detection cycle. The image output range determination unit 120 can determine the final position of the image output range by determining the center position of the image output range as follows, for example.
Upon receiving the feature point position information (referred to as vector S), the image output range determination unit 120 receives the center position (vector P _n , n) of the image output range determined by the image output range determination unit 120 in the previous detection cycle. Calculates the output range shift amount (| SP _n |) from the detection cycle number). When the output range moving amount exceeds a predetermined value (M), the next image center position (tentatively, the vector P) is set at an arbitrary position between the feature point position information and the center of the currently set image output range. _{n + 1} ). When the output range movement amount does not exceed the predetermined value M, the image output range is determined around the feature point position information of the feature points.
For example,
When | S-P _n |> M, P _{n + 1} = P _n + (S−P _n ) · (M / | S−P _n |)
When | S−P _n | ≦ M, P _{n + 1} = S
Thus, the image center position is determined. By performing such processing, it is possible to output the output image while moving the image output range at a predetermined speed so as to follow the instantaneously moved feature point.
Alternatively, as described below, a multiple of the predetermined value M may be added as a reference value to set the center position of the image output range.
When | S−P _n |> 2M, P _{n + 1} = P _n + (S−P _n ) · (M / | S−P _n |)
When 2M ≧ | S−P _n |> M, P _{n + 1} = P _n + (S−P _n ) / 2
When | S−P _n | ≦ M, P _{n + 1} = S

The camera device 500 described in this embodiment includes:
The feature point detection unit 130 further stores the feature point position information of the detected feature points for each detection cycle,
The feature point detection unit 130 calculates the feature point movement amount from the feature point position information of the detected feature point and the stored feature point position information of the feature point detected one cycle before each feature period. Calculated as travel,
The image output range determination unit 120 determines an output range according to the feature point movement amount.

The camera device 500 described in this embodiment includes:
The image output range determination unit 120 further stores the determined output range for each detection cycle,
The amount of movement of the output range is calculated as the amount of movement of the output range from the feature point position information of the feature point detected by the feature point detection unit 130 and the stored center position of the output range determined one cycle before. When the output range movement amount exceeds a predetermined value, the output range is determined according to the output range movement amount.

According to the camera device 500 described in the present embodiment, since the feature point detection unit 130 calculates and stores the feature point movement amount, the image output range determination unit 120 includes the feature point type information, the feature point position. It is possible to determine the image output range together with the information and the feature point movement amount.
Further, the image output range determination unit 120 stores the image output range for each detection cycle. For this reason, when the feature point moves greatly instantaneously, the image output range determination unit 120 can move the center at a predetermined speed instead of instantaneously moving the center of the image output range. As a result, even if the feature point moves instantaneously, the image is not instantaneously switched, and an easy-to-see image can be output to the display device or the recording device.

Embodiment 4 FIG.
In the present embodiment, the operation of the camera device 500 when the feature point detection unit 130 detects a plurality of feature points from a wide range image captured by the imaging unit 200 will be described.

  When there are a plurality of feature points in a wide range image captured by the image capturing unit 200, if the image output range determination unit 120 allows all the feature points to fall within the image output range, the image output range becomes wide and detailed images cannot be output. End up.

  Therefore, the camera device 500 according to the present embodiment has not only the position of the feature point in the image identification circuit 100 but also its type and detection status (the amount of motion in the case of motion detection, the captured face in the case of a face). And a function for determining a range for outputting an image according to an arbitrary rule from a generated temporal relationship.

For example, when motion is detected at both ends of the shooting range of a security camera equipped with a wide-angle lens, the image with the larger motion is selected and output. If this state continues, the image is output after switching to the other point after a few seconds. It is possible to output detailed images of feature points having two movements.
Alternatively, after a face is detected from a wide-range image and the feature point is set as an image output range, if a motion is detected at another feature point, the image is immediately switched and the image is output. If an image of a person's face is output even for a moment, an image of the person's face can be left in the recording device 920. Therefore, a necessary image can be obtained by immediately enlarging a feature point having the next movement.

  A configuration of camera device 500 according to the present embodiment will be described with reference to FIG. FIG. 6 is a block configuration diagram showing a configuration of a recording system including the camera device 500 according to the present embodiment.

  The feature point detection unit 130 of the camera device 500 according to the present embodiment is divided into a feature point N detection unit 130-1 and a feature point N detection unit 130-N in order to detect a plurality of feature points. For example, the feature point 1 detection unit 130-1 detects a moving object as the feature point 1, and the feature point 2 detection unit 130-2 is a feature point at a place other than that detected by the feature point 1 detection unit 130-1. 2 detects a human face.

Each feature point detection unit 130 simultaneously detects feature points from a wide-range image, and sends the degree of detection as a feature point detection amount to the image output range determination unit 120. For example, the feature point detection unit 130-1 detects a feature point, determines each position (feature point position information), an amount of movement (feature point movement amount), and feature point type information, and uses it as a feature point detection amount. The image is sent to the image output range determination unit 120. The image output range determination unit 120 determines the image output range based on a predetermined rule using all the feature point detection amounts sent from the respective feature point detection units 130.

  The predetermined rule is, for example, when the feature point detection unit 130 detects a plurality of feature points and each feature point detection amount is input to the image output range determination unit 120. This is a rule that determines whether to detect feature points detected in order or how much time to take each feature point. This rule is recorded in advance as a file, and the image output range determination unit 120 determines the image output range with reference to the file in which this rule is recorded. The image output range determination unit 120 outputs a plurality of feature points detected by the feature point detection unit 130 to the recording device 920 and the display device 400 in a clear state, and therefore determines the priority order when determining the image output range. To do. The image output range determination unit 120 stores the priority determined by itself and the information of the number of times and the time when the image output range is determined for each feature point while the feature point detection unit 130 detects the same feature point. It shall be.

For example, when the feature point detection unit 130 detects three feature points from a wide range image captured by the imaging unit 200, the image output range determination unit 120 operates as follows.
As a specific example of the predetermined rule, for example, the priority when determining is “person, luggage, unknown”, and the display time (the time for which the display device displays the output image) is set to “one cycle based on the detection cycle. 2 cycles, 5 cycles ". When the movement of the feature point is detected, each display time is doubled.
The operation is shown below. First, the feature point detection unit 130 detects feature points. Among the detected feature points, the feature point type information of each feature point is that feature point 1 is a human face, feature point 2 is a load, and feature point 3 is unknown. It is assumed that the feature point movement amount of the feature point 1 and the feature point 2 is substantially 0, but the feature point 3 frequently moves.
After detecting each feature point, the feature point detection unit 130 inputs each feature point position information, feature point type information, and feature point movement amount to the image output range determination unit 120.
The image output range determination unit 120 determines the priority order, output time, and number of times for determining the image output range for each feature point based on a predetermined rule. According to the above-mentioned predetermined rule, since an image is output with the highest priority on humans, first, an image output range is determined for the feature point 1. Since a human face is sufficient for recording if it is output for a moment, it is displayed until one detection cycle has elapsed. Since the next priority is the package, the image output range for the feature point 2 detected as the package is determined. The feature point 2 is imaged for two detection periods according to a predetermined rule. The image output unit 300 outputs an output image displaying the feature point 2 to the recording device 920 or the display device 400.
After two cycles, the image output range determination unit 120 determines the image output range for the feature point 3 that is the next priority. The feature point 3 cannot be detected, but there is movement. Therefore, the image output range determination unit 120 determines the image output range for the feature point 3 for 10 cycles, which is a double of 5 cycles. The unknown feature point is displayed on the display device 400 for a long time compared to other feature points, and the user of the camera device 500 can confirm what the feature point 3 is from the movement of the feature point 3. . For example, it can be recognized from the movement of the feature point 3 that the feature point 3 is an insect.

The camera device 500 described in this embodiment includes:
When the feature point detection unit 130 detects at least two or more feature points from the wide range image input by the photographing unit 200 and detects a plurality of feature points from the wide range image input by the photographing unit 200, each feature point In addition, the feature point type information, the feature point position information, and the feature point movement amount are output to the image output range determination unit 120,
The image output range determination unit 120 is characterized in that an output range is determined according to a predetermined rule from feature point type information, feature point position information, and feature point movement amount of a plurality of feature points.

  According to the camera device 500 of the present embodiment, a plurality of feature points are detected, the image output range determination unit 120 determines an image output range for each feature point, sets the order, and outputs the result. There is an effect that it is possible to output a detailed image for each detected feature point.

Embodiment 5 FIG.
In the present embodiment, an installation location (attachment position) when the camera device 500 described in the first to fourth embodiments is used as a security camera device for an elevator will be described.
7A and 7B are diagrams for explaining the attachment position of the camera device 500. FIG. 7A is a view of the attachment position of the camera device 500 as seen from inside the elevator car, and FIG. 7B is the camera device. It is a top view which shows the attachment position of 500. FIG.

A mounting position of the camera device 500 will be described. The camera device 500 can photograph the entire interior of the elevator car by installing the photographing unit 200 including the wide-angle lens on the ceiling in the elevator car. An image in the elevator car photographed by the photographing unit 200 through the wide angle lens can be taken into the camera device 500 as a wide range image.
Further, as shown in FIG. 7, the camera device 500 is, for example, near the door in the car (near the door side of the car door), on the lower side of the upper plate 710 where the door is closed, particularly at the position where the door is closed. Install. By attaching the camera in this position so that the shooting direction of the camera is directly below (or in the vicinity thereof), not only the inside of the elevator car but also the door image can be taken when the door is opened. In this case, the angle of view of the wide-angle lens used in the camera device 500 is assumed to be close to 180 degrees.

The photographing unit 200 of the camera device 500 described in the present embodiment has a photographing direction in which the inside of the elevator car is photographed downward near the ceiling or the upper plate of a portion where the door doors of the elevator car meet each other. It is attached.
For this reason, when the door is opened, not only the inside of the elevator car but also the door is opened, it is possible to take an image of the landing.

Embodiment 6 FIG.
A camera device 500 according to the present embodiment will be described with reference to FIGS.
FIG. 8 is a block diagram showing the structure of an elevator crime prevention system 1100 that uses the camera device 500 according to the present embodiment. FIG. 9 is a diagram showing a detection range of the door sandwiched by the camera device 500. FIG. 9A is a view in which the door is sandwiched and the detection range is viewed from inside the elevator, and FIG. 9B is a plan view. FIG. 10 is a diagram illustrating a detection range of the door being retracted detected by the camera device 500. FIG. 10A is a view in which the door is retracted and the detection range is viewed from inside the elevator, and FIG. 10B is a plan view thereof.

  In the following description, differences between the present embodiment and the first to fifth embodiments will be mainly described. In FIG. 8, the same numbers are given to functional blocks indicating the same functions and operations shown in the first to fourth embodiments, and the description thereof is omitted. In FIG. 8, the difference from the first embodiment is that the camera device 500 includes a detection unit 150, a door retract detection unit 160, and a signal output unit 170 when the door is caught. The elevator crime prevention system 1100 also includes an elevator control device 600 that inputs a signal output from the signal output unit 170 and controls the opening and closing of the door 700 and the elevator operation according to the signal. The destination from which the signal output unit 170 outputs a signal may be a door control device that controls opening and closing of the door door 700. The elevator control device 600 described below includes a door control device.

The door pinching detection unit 150 analyzes a wide range image input from the photographing unit 200 when the door door 700 is closed, and detects a feature point included in the door pinching detection range 810. When the door pinching detection unit 150 detects a feature point (an object pinched by the door door 700), it outputs a signal indicating the detection to the signal output unit 170. Also, a signal is output even if motion is detected from an image near the door on the landing side.
When a signal indicating that there is a feature point between the door doors 700 in the elevator car is input from the door pinching detection unit 150, the signal output unit 170 is input to the elevator controller 600 to the door in the elevator car. A detection signal is output between the doors indicating that there is a feature point between them. The signal output from the signal output unit 170 is transmitted to the elevator control device 600 by, for example, serial communication or a contact signal.
The door pinching detection range 810 is a range located between the door doors 700 as shown in FIG. When the door 700 is open, the camera device 500 takes a picture including the landing part.
When the detection signal is input from the signal output unit 170, the elevator control device 600 temporarily stops the opening / closing operation of the door door 700 (in particular, stops the operation of closing the door door 700). The elevator control device 600 resumes opening and closing of the door 700 after the door is pinched from the signal output unit 170 and the detection signal disappears and safety is confirmed.

When the door is closed from the image (wide range image) photographed by the photographing unit 200, the door retracted detection unit 160 is in the vicinity of the door door 700 and the door pocket 720 (the door retracted detection range 820) immediately before the door is opened or in the middle of opening the door. A feature point (an object drawn into the door 700) is detected, and a signal indicating the detection is output.
When the signal output unit 170 receives a signal from the door retracted detection unit 160 that detects that the door is retracted and has a characteristic point in the detection range, the signal output unit 170 causes the elevator control device 600 to send the door door 700 in the elevator car. And a door pull-in signal indicating that there is a feature point in the vicinity of the door pocket 720.
When the door is drawn and a signal is input from the signal output unit 170, the elevator control device 600 temporarily stops the opening operation of the door door 700, for example. The elevator control device 600 resumes the opening and closing of the door 700 after the door is pulled in from the signal output unit 170 and the detection signal disappears and safety is confirmed.
As shown in FIG. 10, the door retracted detection range 820 is a range between the door door 700 and the door pocket 720 when viewed from within the elevator car.

The camera device 500 described in this embodiment includes:
A door pinching detection unit 150 for detecting feature points located between open doors in an elevator car from a wide range image captured and input by the imaging unit 200;
When the door pinching detection unit 150 detects a feature point located between the open doors in the elevator car, a door pinching detection signal indicating that there is a feature point between the doors in the elevator car is provided. And a signal output unit 170 for outputting.

The camera device 500 described in the present embodiment also detects a door pull-in detection that detects a feature point located in the vicinity of a door and a door in an elevator car from a wide range image captured and input by the imaging unit 200. Part 160;
When the door pull-in detection unit 160 detects a feature point located in the vicinity of the door between the door and the door in the elevator car, it indicates that there is a feature point near the door in the elevator car and the door bag. And a signal output unit 170 that is retracted and outputs a detection signal.

  If the elevator crime prevention system 1100 including the camera device 500 described in this embodiment is used, feature points between doors or near door door pockets are detected, and when feature points are detected, the signal output unit 170 is used. The signal is output to the elevator control device 600. Therefore, there is an effect that the camera device 500 can commonly perform detection of the door being pinched and detection of the door being retracted, which is normally detected by a dedicated sensor.

  Of the embodiments described above, two or more embodiments may be combined. Alternatively, one of these embodiments may be partially implemented. Or you may implement combining two or more embodiment among these partially.

  In addition, the flowcharts of FIGS. 2 and 5 described in each embodiment correspond to a processing procedure of a camera control program executed on a computer that realizes the camera device 500.

1 is a block configuration diagram illustrating a configuration of a recording system 1000 including a camera device 500 according to Embodiment 1. FIG. 2 is a diagram illustrating an example of a hardware configuration of a camera device 500 and a recording system 1000 according to Embodiment 1. FIG. 5 is a flowchart showing the operation of the camera device 500 according to Embodiment 1. 6 is a block configuration diagram illustrating a configuration of a recording system 1000 including a camera device 500 according to Embodiment 2. FIG. 6 is a flowchart illustrating an operation of the camera device 500 according to Embodiment 2. FIG. 10 is a block configuration diagram illustrating a configuration of a recording system 1000 including a camera device 500 according to Embodiment 4. It is a figure which shows the attachment position of the camera apparatus 500 which concerns on Embodiment 5. FIG. It is a block diagram which shows the structure of the elevator crime prevention system 1100 containing the camera apparatus 500 which concerns on Embodiment 6. FIG. It is a figure which shows the door pinching detection range 810 which the camera apparatus 500 detects. It is a figure which shows the door drawing-in detection range 820 which the camera apparatus 500 detects.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 Image identification circuit, 110 Image extraction part, 120 Image output range determination part, 130 Feature point detection part, 140 Image distortion correction part, 150 Door pinching detection part, 160 Door retracted detection part, 170 Signal output part, 180 Timer , 200 photographing unit, 300 image output unit, 400 display device, 600 elevator control device, 700 door door, 710 upper plate, 720 door bag, 810 door pinched detection range, 820 door retracted detection range, 911 CPU, 913 ROM, 914 RAM, 902 input device, 915 communication board, 920 recording device, 921 operating system, 922 program group, 923 file group, 1000 recording system, 1100 elevator security system.

Claims (11)

A photographing unit that photographs a predetermined range and inputs an image of the photographed predetermined range as a wide-range image;
An image processing unit for processing the wide range image input by the imaging unit;
An image output unit that outputs at least a part of the wide range image processed by the image processing unit as an output image;
The image processing unit
Feature points indicating features of the wide range image are detected from the wide range image input by the photographing unit by a predetermined detection procedure, and feature point type information indicating the type of the feature point based on the detection result and the wide range image A feature point detection unit that determines the feature point position information indicating the position of the feature point in and outputs the determined feature point type information and the feature point position information;
An image output range determination unit that inputs the feature point type information and the feature point position information, and determines an output range by a predetermined range determination procedure using the feature point type information and the feature point position information;
An image extraction unit that extracts an image included in the output range determined by the image output range determination unit as an extracted image from the wide range image input by the imaging unit, and outputs the extracted image to the image output unit; Prepared,
The image output unit includes:
The camera apparatus, wherein the extracted image extracted by the image extraction unit is output as the output image. The camera device further includes:
A timer for measuring an imaging cycle indicating an interval at which the imaging unit captures an image, and a detection cycle indicating an interval at which the feature point detection unit detects the feature point;
The photographing unit
For each shooting period measured by the timer, a predetermined range is shot, and an image of the shot predetermined range is input as the wide range image,
The feature point detector
A feature point indicating the feature point of the wide-range image input from the photographing unit is detected by a predetermined detection procedure for each detection period measured by the timer, and the feature point type is indicated based on the detection result. Output point type information and feature point position information indicating the position of the feature point in the wide range image,
The image output range determination unit
For each detection cycle, the feature point type information and the feature point position information are input, an output range is determined by a predetermined range determination procedure using the feature point type information and the feature point position information,
The image extraction unit
For each detection period, image information included in the output range determined by the image output range determination unit is extracted as an extracted image from the wide range image input by the imaging unit, and the extracted image is input to the image output unit. The camera device according to claim 1, wherein the camera device outputs. The feature point detection unit further includes:
Storing feature point position information of detected feature points for each detection period;
For each detection period, the amount of movement of the feature point is calculated as the amount of movement of the feature point from the feature point position information of the detected feature point and the stored feature point position information of the feature point detected one cycle before, Output to the image output range determination unit,
The image output range determination unit receives the feature point movement amount, and determines an output range according to the feature point type information, the feature point position information, and the feature point movement amount. 2. The camera device according to 2. The image output range determination unit further includes:
For each detection cycle, store the determined output range,
From the feature point position information of the feature point detected by the feature point detection unit and the stored center position of the output range determined one cycle before, the movement amount of the output range is calculated as the output range movement amount,
The camera apparatus according to claim 3, wherein when the calculated output range movement amount exceeds a predetermined value, an output range is determined according to the output range movement amount. The feature point detector
Detecting at least two or more feature points from the wide range image input by the photographing unit;
When detecting a plurality of feature points from the wide range image input by the imaging unit,
For each feature point, feature point type information, feature point position information, and feature point movement amount are output to the image output range determination unit,
The image output range determination unit
5. The camera apparatus according to claim 3, wherein an output range is determined according to a predetermined rule from feature point type information, feature point position information, and feature point movement amount of the plurality of feature points. The photographing unit
Shooting the predetermined range at a wide angle using a wide angle lens,
The image processing unit further includes:
The extracted image is input from the image extracting unit, the distortion of the extracted image caused by the photographing unit photographing the predetermined range at a wide angle using the wide angle lens is corrected, and the corrected image is corrected. As an image distortion correction unit that outputs to the image output unit,
The image output unit includes:
The camera apparatus according to claim 1, wherein the correction image is input from the image distortion correction unit and output as the output image. The camera device is
The camera device according to claim 1, wherein the camera device is installed in an elevator car and images the inside of the elevator car as the predetermined range. The photographing unit
8. The camera device according to claim 7, wherein the camera device is attached to a portion near a ceiling or an upper plate of a portion where door doors in the elevator car meet each other with a photographing direction for photographing the inside of the elevator car facing downward. . The camera device is
A door pinching detection unit for detecting a feature point located between open doors in the elevator car from the wide range image photographed and input by the photographing unit, and the door pinching detection unit includes the elevator car A signal output unit that outputs a detection signal between the doors indicating that there is a feature point between the doors in the elevator car when a feature point located between the open doors is detected. 8. The camera apparatus according to claim 7, wherein The camera device is
A door retract detection unit that detects a feature point located in the vicinity of a door and a door in the elevator car from the wide range image captured and input by the image capturing unit, and the door retract detection unit includes the elevator When a feature point located between a door and a door pocket in the car is detected, a door pull-in detection signal indicating that there is a feature point between the door and the door bag in the elevator car is output. The camera apparatus according to claim 7, further comprising a signal output unit. A camera device according to any one of claims 1 to 10;
A recording device for recording the output image output by the image output unit of the camera device;
A recording system comprising: a display device that displays the output image.

Images