KR20120038640A - Safty system of heavy weight moving apparatus using image processing - Google Patents

Abstract

PURPOSE: A security system of a load transport device using image processing is provided to recognize an operator's protective helmet through an image processing method by the installation of a video camera in the load transport device. CONSTITUTION: A video camera(10) is installed in a load transport device(1). An image processing device(20) distinguishes an operator by analysis of the images recorded by the video camera. The image processing device controls alarm by determining the position and distance of a discriminated operator. An operator monitoring panel(30) is installed in an inner driver's seat of the load transport device. The operator monitoring panel receives the drive-operating signals through a touch panel. An alarm device(40) generates alarms and warning light with the alarm control of the image processing device to the operator and an external worker.

Description

Safety system of heavy weight moving device using image processing {SAFTY SYSTEM OF HEAVY WEIGHT MOVING APPARATUS USING IMAGE PROCESSING}

The present invention relates to a safety system of a heavy weight moving apparatus using image processing, and more particularly, to install an image camera on a heavy weight moving apparatus to acquire an image, and detect a safety helmet of an operator from the acquired image to be within a safe distance. The present invention relates to a safety system of a heavy weight moving apparatus using image processing that operates a warning light and generates a warning sound to a driver, thereby preventing a safety accident in advance.

In shipyards in which ships are built, most of the conveyed objects are heavy objects, and heavy objects are moved by using heavy goods moving devices such as forklifts and crane transformers.

There is a high risk of safety accident as there may be other workers around the moving distance when moving heavy goods using heavy goods moving device. Accordingly, in many cases, an auxiliary worker for guiding the movement of the heavy weight moving device from the rear of the heavy weight moving device is required.

The present invention is equipped with a camera in the heavy weight moving apparatus to acquire an image, and automatically identifies the operator's helmet by the acquired image processing to warn if it is within the set proximity distance to prevent the accident in advance to prevent the accident It is to provide a safety system of the heavy weight moving device used.

In addition, the present invention is to provide a rear image photographed to the driver so that the driver can directly identify the rear obstruction, and to attach the marker to the hard hat to easily determine the presence of the operator.

An object of the present invention is installed in the heavy weight moving device and the image camera for taking images of the side and the back of the heavy weight moving device;

An image processing apparatus for identifying a worker by analyzing the captured image of the image camera, determining a distance of the identified worker, and controlling an occurrence of an alarm when within a preset distance;

It is installed in the driver's seat inside the heavy weight moving device to display the image taken by the camera on the screen under the control of the image processing device and output a warning sound according to the occurrence of a warning, and receive a driving operation signal through the touch panel. A driver monitoring panel;

And an alarm device for generating a warning sound and a warning light to an external worker by warning generation control of the image processing device.

The present invention is a method for identifying the worker, using a hard hat worn by the worker, identification means attached to the work clothes, belongings, etc., may use a marker.

The present invention is characterized in that at least one or more cameras are provided, and the driver monitoring panel displays images of each camera by dividing the screen into multiple screens. That is, since the heavy object moving device can move forward and backward, the camera attached to the heavy object moving device is installed in a number capable of photographing the surroundings without a blind spot.

The present invention is characterized in that the camera is a camera that can shoot day and night. That is, there is a feature in using a camera that includes an infrared camera function.

The present invention is characterized in that a marker for image identification is further attached to the helmet of the operator so that the operator can be clearly identified when shooting the camera.

The marker of the helmet is characterized in that the fluorescent tape attached to the helmet in the form of a border.

The present invention has a high risk of a safety accident due to the weak rear monitoring when the heavy weight moving device is moved in a place where there are many heavy weight movements and workplaces, such as a shipyard, to install a video camera on the heavy weight moving device, and to use an image processing method. Object recognition of the worker's hard hat, and if the distance and position of the recognized hard hat within the preset warning range to warn the driver and external workers, it is possible to prevent the safety accident in advance.

1 is a schematic diagram of a safety system of a heavy weight conveying apparatus according to the present invention;
2 is a configuration of the safety system of the heavy weight conveying apparatus according to the present invention.
3 is a detailed configuration diagram of an image processing apparatus according to the present invention.
4 is a diagram illustrating a scene in which a human hand is extracted using an image processing algorithm according to the present invention.

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

1 is a schematic diagram of a safety system of a heavy weight conveying apparatus according to the present invention, and FIG. 2 is a configuration of a safety system of a heavy weight conveying apparatus according to the present invention.

An image camera 10 installed at the heavy weight moving device 1 and taking images of the driver's side and back;

An image processing apparatus (20) for analyzing a photographed image of the image camera (10) to identify a worker, determining a location and a distance of the identified worker, and controlling an occurrence of a warning when within a preset distance;

It is installed in the driver's seat inside the heavy weight moving device 10 to display the image taken by the image camera 10 on the screen under the control of the image processing device 20 and output a warning sound according to the warning occurrence, A driver monitoring panel 30 receiving a driving operation signal through a touch panel;

It is characterized in that it comprises an alarm device 40 for generating a warning sound and warning lights to the driver and external workers by the alarm generation control of the image processing apparatus 20.

According to the present invention, at least one or two or more video cameras 10 are installed, and the driver monitoring panel 30 displays images of the video cameras 10 by dividing them into multiple screens. do.

The present invention is characterized in that the video camera 10 is a camera capable of day and night shooting, including the infrared camera function.

The present invention can identify the worker's hard hat (2) to clearly identify the worker when shooting the video camera 10, or may use a safety belt or a separate color identification means or belongings attached to the worker's work clothes, In the embodiment of the present invention, the identification of the hard hat 2 will be described as an example.

In addition, an image identification marker is further attached to a safety helmet or work clothes of the worker, and the image identification marker is characterized in that a color identification marker is attached to the camera for easy color identification during image processing.

When the marker is attached to the hard hat (2) is characterized in that the fluorescent tape attached to the helmet in the form of a border.

3 is a detailed block diagram of an image processing apparatus according to the present invention.

The image processing apparatus 20 according to the present invention includes a filtering unit 21 for receiving and filtering an image signal photographed from the image camera 10, and a safety helmet 2 in the image filtered by the filtering unit 21. A color model converter 22 for converting a color corresponding to an image from a color of another object, and a pixel of an image obtained by color model conversion from an image output from the color model converter 22. To extract the image to the helmet 2 to extract the image, and to perform the interpolation process using a morphology for the image output from the specific pixel extraction unit 23 to process the image interpolation for the helmet An interpolation unit 24 using a morphology, an object extraction unit 25 for extracting an image of a hard hat by an object extraction method by labeling from the image interpolated by the interpolation unit 24, and the object extraction unit 25. In Preset warning of distance and position of the helmet recognized by the distance position recognition unit 26 and the distance position recognition unit 26 of the object recognizing distance and position information according to the size of the object to the helmet The alarm generating unit 27 is configured to warn and notify the driver and the external worker through the alarm device 40 and the driver monitoring panel 30 within the range of the occurrence distance and the position.

According to the present invention constituted as described above, the image camera 10 is installed in the heavy weight conveying apparatus 1, and the image camera 10 captures the side and rear side images of the heavy weight moving apparatus 1, and the worker's helmet 2 Be aware of this. The reason for the worker's hard hat 2 to be recognized is that the hard hat 2 must be worn at the work site, and the hard hat 2 is clearly identified from other objects and can be easily identified by image processing. Therefore, in the present invention, the hard hat 2 is identified as an example. In addition to the hard hat, if the worker's safety belt or a separate marker is to be worn, the safety belt or marker can be identified by image processing.

It is preferable to install at least two or more video cameras 10. In the case of a large heavy weight moving device such as a transformer, the radius to be monitored is large, and in the case of a forklift, it is possible to rapidly rotate even in a very narrow space. It is preferable to install two or more because it is difficult to determine the presence or absence of the worker behind the side by using only the video camera. That is, since the heavy weight moving device is movable back and forth, the camera attached to the heavy weight moving device is installed in a number capable of photographing the surroundings without a blind spot.

When two or more image cameras 10 are installed, the image processing apparatus 20 is configured in parallel to match the number of cameras so as to perform image processing for each camera.

In addition, the image camera 10 uses a camera capable of day and night shooting, including an infrared camera function, so that the image can be clearly captured even in a dark workplace or at night.

The image processing apparatus 20 according to the present invention receives an image photographed from the image camera 10 and filters it by the filtering unit 21, and converts the color model by the color model converting unit 22. When the color model is converted, a specific pixel extraction unit 23 can extract the desired color, and the interpolation unit 24 performs interpolation using a morphology. Then, the object extraction unit 25 performs object extraction through labeling, and the object distance distance recognition unit 26 recognizes the distance and position of the object with respect to the hard hat. Such an image processing algorithm may be implemented by a well-known conventional image processing technique, which is not described in detail in the present invention. However, the present invention is characterized by using a color model conversion method as a method for recognizing a hard hat.

4 is an exemplary view of extracting a human hand using an image processing algorithm according to the present invention. The picture at the top shows a small circular dot as shown on the right when the hand is extracted when only the palm is visible. The picture below shows the hand of a person even if the part of the person's neck or the other arm is present. It is a scene to extract. As such, it can be seen that even a hand that does not have a constant color (although it is called a hand but actually has a slightly different color for each part) and a hand that does not have a constant shape, can be extracted well.

Here, in the embodiment of the present disclosure, the image of the worker's helmet 2 is identified by color model conversion. However, the present invention is not limited thereto. It can be distinguished from work clothes and surrounding objects, and workers can be identified by color model conversion without any means. In order to identify the worker more clearly, the worker may wear the work cap 2 or the safety belt or an image identification marker (for example, a fluorescent cloth or a band capable of color identification) attached to the work clothes, and the worker by image processing. Can be identified.

Therefore, the recognition rate of the object having a certain shape and color, such as a hard hat, is further increased. And if you use a marker on the helmet, the recognition rate can be even higher.

1 is an example of applying a safety system of a heavy weight moving apparatus using the image processing of the present invention to a forklift. By analyzing the image in real time through the image of the camera as shown in Figure 1 if there is an operator within the dangerous distance of the forklift is a method of alerting both the external worker and the forklift driver inside. Because this is a system that is specific to a particular task, you may be able to use a system that matches an LCD and a small computer, rather than a larger PC like a regular desktop.

1: heavy weight moving device 2: hard hat
10: video camera 20: image processing device
21: filtering unit 22: color model conversion unit
23: specific pixel extraction unit 24: interpolation unit
25: object extraction unit 26: distance position recognition unit of the object
27: alarm generating unit 30: driver monitoring panel
40: alarm device

Claims (7)

An image camera (10) installed in the heavy object moving device (1) for capturing images of the side and the rear of the heavy object moving device (1);
An image processing apparatus (20) for analyzing a photographed image of the image camera (10) to identify a worker, determining a location and a distance of the identified worker, and controlling an occurrence of a warning when within a preset distance;
It is installed in the driver's seat inside the heavy weight moving device 10 to display the image taken by the image camera 10 on the screen under the control of the image processing device 20 and output a warning sound according to the warning occurrence, A driver monitoring panel 30 receiving a driving operation signal through a touch panel;
Safety system of the heavy object moving apparatus using the image processing, characterized in that made by the alarm processing control to generate a warning sound and warning lights to the driver and the external worker by the alarm generation control of the image processing device (20).
The image processing apparatus 20 of claim 1,
Safety system of heavy goods moving apparatus using image processing, characterized in that the identification of the worker by the image analysis process by the color model conversion of the hard hat image of the worker in order to identify the worker by the image taken by the image camera.
The method according to claim 1, wherein the image processing apparatus 20 attaches an image identification marker to the worker's work clothes or hard hat 2 to identify the worker by image analysis by color model conversion. Safety system of heavy weight moving device using image processing.
The safety system of claim 3, wherein the image identification marker is a fluorescent tape attached to a hard hat in a frame shape.
The image of claim 1, wherein two or more video cameras 10 are installed, and the driver monitoring panel 30 displays images of the video cameras 10 by dividing them into multiple screens. Safety system of heavy weight moving device using treatment.
The method of claim 1, wherein the video camera 10,
Safety system of heavy goods moving apparatus using image processing, characterized in that the camera capable of day and night shooting, including the infrared camera function.
The image processing apparatus 20 of claim 1,
Filtering unit 21 for receiving and filtering the image signal taken from the image camera 10, and the color corresponding to the image of the helmet (2) in the image filtered by the filtering unit 21 of the color of another object Color model conversion unit 22 for color model conversion to identify and extract the pixels of the color model converted image from the image output from the color model conversion unit 22 to extract the image to the helmet 2 An interpolation unit 24 using a morphology that performs interpolation using a morphology on an image output from the specific pixel extraction unit 23 and the specific pixel extraction unit 23, and the image interpolation for a hard hat; The object extractor 25 extracts the image of the hard hat by the object extraction method through the labeling from the interpolated part 24 and the size of the object for the hard hat extracted by the object extractor 25. The distance position recognition unit 26 of the object recognizing the distance and position information according to the distance, and the distance and the position of the hard hat recognized by the distance position recognition unit 26 of the object is within the preset warning distance and position range alarm Safety device for heavy goods moving apparatus using image processing, characterized in that it comprises an alarm generator for warning and notifying the driver and external workers through the device (40) and the driver monitoring panel (30).

Images