KR101805564B1 - Alarm system for prevent ship collision and method thereby - Google Patents

Abstract

Provided are an alarm system for preventing a ship collision and a method thereof. According to an embodiment of the present invention, the alarm system for preventing a ship collision comprises: a photographing unit photographing the outside of a ship, and formed to increase visibility distinguished through fog correction when there is fog; and a ship collision prediction unit receiving an image acquired in the photographing unit to analyze the image and determine whether an object is photographed in the image, and determining danger of a collision with the object to generate an alarm.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an alarm system,

The present invention relates to an alarm system and method for preventing collision of a ship, and more particularly, to a system and method for providing an alarm so as to prevent a collision when a vessel is navigated.

Generally, a ship is a structure that can be floated in water and loaded with people, livestock and materials and can move over water. Vessels can be divided into merchant ships for the purpose of importing the freight, warships for war activities, fishing vessels engaged in fishery, and special ships for the purpose of carrying out special business by transporting passengers or cargoes according to the purpose of use.

If there is a collision that does not recognize obstacles such as reefs and floats in front of them, or does not recognize a ship operating on another route, there will be a lot of human, property, and environmental damage. In order to prevent this, Korean Patent Registration No. 10-1564690 discloses a navigational assistance device and method for identifying a reef or a float present on the front sea of a ship using images and adjusting the operation of the ship do.

However, in the case of such a navigation assistant apparatus and method, when a natural phenomenon occurs in the vicinity of a ship where the visual distance is drastically reduced at the time of capturing an image such as seaweed or fog, the visibility distance visible by the image is drastically reduced, The collection of information on dangerous objects in front of the ship is restricted and there is still a problem that marine accidents may occur.

KR 10-1564690 B

In order to solve the problems of the conventional art as described above, one embodiment of the present invention is to detect and detect a dangerous object in front of the ship even if a change in weather occurs, And to provide a warning system and method for preventing collision of a ship.

According to an aspect of the present invention, there is provided an alarm system for preventing collision with a ship. Wherein the warning system for preventing collision of a ship captures the outside of a ship and is configured to increase the visibility distance that can be identified through fog correction in case of fogging; And a ship collision predicting unit for receiving the image acquired by the photographing unit and analyzing the image to confirm whether or not an object is photographed in the image, and determining a risk of collision with the object to generate an alarm.

Wherein the photographing unit comprises: a camera module for photographing the outside of the ship to acquire the image; A fog correction module for correcting fog in an image acquired through the camera module to increase an identifiable visible distance; And a visible range adjustment module capable of adjusting an enlargement degree of the image to identify the object included in the image.

The fog correction module may remove fog from the image using a Median Dark Channel Prior (MDCP).

The fog correction module may further perform color distortion correction of the image and edge value correction of the image after removing the fog.

When the object is photographed in the image, the visible-distance adjustment module may enlarge the image for a predetermined time using a variable zoom lens.

Wherein the ship collision prediction unit analyzes the image and confirms whether the object is photographed in the image; And a collision determiner for determining whether there is a collision risk that the ship may collide with the object identified by the image processor.

Wherein the image processing unit comprises: An image analysis module for analyzing the image to check whether the object is photographed; An image storage module for storing the analyzed image in the image analysis module; And an object determination module that determines whether the object is a dangerous object that may collide with the ship such as another ship or a reef and may threaten the hull of the ship.

The image analysis module may analyze the image on which the fog correction is performed to confirm whether an object is photographed on the image.

The image analysis module may use a Haar-like feature technique to identify the object.

The image storage module may store the image when the image analysis module determines that the object is not photographed, and may use the stored image as a basic image in which the object is not photographed.

The object determining module may obtain an enlarged image from the photographing unit to determine whether the object is the dangerous object.

The collision determination unit may include a distance measurement module that calculates a distance to a dangerous object, which is an object with which the ship may collide; And a collision determination module that determines whether the dangerous object is in a collision danger distance using the distance calculated by the distance measurement module and determines a collision alarm type.

The distance measurement module measures the distance from the dangerous object using the number of pixels and the pixel distance in the reference image photographed in the actual 10m unit period measured in the period, and the reference image can be photographed at 2 megapixels.

The distance measurement module may further include a speed measurement module for calculating a relative speed between the ship and the dangerous object.

Wherein the collision determination module generates a warning when the distance from the dangerous object calculated by the distance measurement module is longer than the collision danger distance, and when the distance to the dangerous object is closer to the collision danger distance, Can be generated.

The collision determination unit may output the warning alert to the inside of the ship, and output the danger warning to the outside of the ship.

According to an aspect of the present invention, there is provided an alarm method for preventing collision with a ship. The method for predicting a collision risk of a ship using an alarm system for preventing a collision of a ship includes the steps of: capturing an image of the outside of the ship to obtain an image and generating a corrected image; Analyzing the corrected image to determine whether there is a dangerous object that may collide with the ship such as another ship or a reef and may threaten the ship; Calculating a distance between the dangerous object and the vessel; And determining whether the dangerous object is in a collision danger distance, and generating an alarm.

The step of generating the corrected image may further include the steps of: acquiring a fog removal image in which an identifiable visible distance is increased by performing fog correction of the image when fog exists outside the ship; And performing color distortion correction of the fog removed image and edge value correction of the image.

The obtaining of the fog removal image may be performed using the median dark channel flier (MDCP).

The step of analyzing the corrected image may include analyzing the corrected image to check whether the object has been photographed, and analyzing whether the photographed object is photographed using a hypo-random feature technique.

In the step of determining the dangerous object, when the object is photographed, the corrected image may be enlarged for a predetermined period of time to determine whether the object is the dangerous object.

The step of determining the dangerous object may further include storing the corrected image if the corresponding object is not determined to be the dangerous object.

The step of calculating the distance may calculate the distance between the dangerous object and the ship using the number of pixels and the pixel distance in the reference image photographed in the actual 10m unit period measured in the past.

The reference image may be photographed at 2 megapixels.

The step of calculating the distance may further include calculating a relative speed of the dangerous object with respect to the ship.

The generating of the alert may include generating a warning alert inside the ship if the dangerous object is not in the collision danger distance and if the dangerous object is in the collision danger distance, .

The step of generating the alarm may further transmit accident information including at least one of the current position, speed and time of the ship through the communication network when the ship is unable to avoid the dangerous object.

An alarm system and method for preventing collision of a ship according to an embodiment of the present invention is capable of easily confirming an object likely to collide with a ship existing in front of the ship even in a sea with mist or seawater.

In addition, an alarm system and method for preventing collision of a ship according to an embodiment of the present invention can reduce the installation cost by using an image camera.

In addition, the alarm system and method for preventing collision of a ship according to an embodiment of the present invention provides different alarms according to the distance to the ship, thereby enabling the position of the ship to be detected even outside and preventing collision.

In addition, according to an embodiment of the present invention, when a collision is inevitable, the ship collision prevention alarm system and method can promptly respond to maritime accidents, marine accidents, This has the effect of drastically reducing human, material, and environmental damage.

1 is a block diagram briefly showing an alarm system for preventing collision of a ship according to an embodiment of the present invention.
2 is a block diagram showing the photographing unit of FIG. 1 in more detail.
3 is a block diagram showing the ship collision prediction unit of FIG. 1 in more detail.
4 is a flowchart briefly showing an alarm method for preventing collision of a ship using an alarm system for preventing collision of a ship according to an embodiment of the present invention.
Fig. 5 is a flowchart showing the process of processing the image of Fig. 4 in more detail.
FIG. 6 is a flowchart illustrating in more detail the step of predicting whether or not the collision shown in FIG. 4 and the step of providing the notification.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art to which the present invention pertains. The present invention may be embodied in many different forms and is not limited to the embodiments described herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and the same or similar components are denoted by the same reference numerals throughout the specification.

FIG. 1 is a block diagram briefly showing an alarm system for preventing collision of a ship according to an embodiment of the present invention. FIG. 2 is a block diagram showing the photographing unit of FIG. 1 in more detail. Fig. Hereinafter, an alarm system for preventing collision of a ship according to an embodiment of the present invention will be described in detail with reference to FIG. 1 to FIG.

The warning system 100 for preventing collision of a ship according to an embodiment of the present invention captures an image of the outside of the ship to acquire an image of the ship, calibrates and analyzes the acquired image, And then generates an alarm selectively according to the distance between the object and the ship. For this purpose, the ship collision warning system 100 may include a photographing unit 110 and a ship collision prediction unit 130.

The photographing unit 110 photographs the outside of the ship and photographs the corrected image in which the visibility distance that can be recognized by the fog correction is increased when the weather for reducing the visible distance such as fog or seawater appears around the ship. The photographing unit 110 may include a camera module 210, a fog correction module 230, and a visible range adjustment module 250.

The camera module 210 photographs the outside of the ship to acquire an external image. However, the present invention is not limited thereto, and the camera module 210 may be provided with one or more camera modules 210 for photographing various directions according to the user's needs. have. At this time, the camera module 210 may have a resolution of 2 million pixels, for example.

The fog correction module 230 corrects fog in the acquired image. The fog correction module 230 may generate a corrected image by removing the fog effect from the image if the external image acquired by the camera module 210 includes fog.

At this time, the fog correction module 230 may be set to operate when it is determined that fog occurs when the visual distance decreases to a certain distance or less, or may be operated by the user directly controlling on / off. ), And may be set to operate when there is hardly any difference in brightness between images.

The fog correction module 230 may also use the MDCP algorithm, which is named Median Dark Channel Prior to remove fog from the external image. Using this, the fog correction module 230 can remove fog in a way that improves the contrast of the fog portion pixels in the RGB color space.

On the other hand, when the fog correction module 230 removes the fog using the MDCP algorithm, the contrast of the fog portion is forcibly improved, so that color distortion and edge value loss in the fog portion may occur. To correct this, the fog correction module 230 can further correct for color distortion and edge value loss that occur after removing the fog. For this purpose, the color accuracy of the fog image is measured in the HIS color space, An algorithm may be used to perform correction for distortion and edge value loss.

The visible range adjustment module 250 may enlarge the image. The visible range adjustment module 250 may be provided to adjust the degree of enlargement of an image in order to more clearly confirm the object in the presence of an object in the image. The visible distance adjustment module 250 may be, for example, a variable zoom lens, and may enlarge an image for a predetermined time to identify an object.

Meanwhile, the correction image obtained by the shooting unit 110 and corrected for the fog is transmitted to the ship collision prediction unit 130. The ship collision prediction unit 130 determines whether there is an object photographed in the correction image, Of the collision risk. For this, the ship collision prediction unit 130 may include an image processor 131 and a collision determiner 133.

The image processing unit 131 is provided to acquire a corrected image and check whether the corrected image has a moving object. In this case, the moving object is not only an object having its own speed but also an object having a relative speed according to the operation of the ship.

In addition, the image processing unit 131 may confirm whether the moving object is an object having a risk of collision with the ship or a dangerous object having a risk of causing a marine accident through collision with the ship. For this, the image processing unit 131 may include an image analysis module 311, an object determination module 313, and an image storage module 315.

The image analysis module 311 analyzes the corrected image and determines whether an object is photographed in the image. The image analysis module 311 can detect an object in the correction image using the machine learning and Haar-like feature technique, and generate an analysis image.

That is, the image analysis module 311 repeatedly learns the detection method using the brightness difference between the area and the area of the lower-level feature technique through machine learning so that the natural phenomena such as waves, waves, rain, It may be judged as noise and removed, and a reef or other ship may be detected as an object in the image. In addition, the image analysis module 311 may further increase the object detection capability using an Adaboost algorithm as needed.

The object determination module 313 determines whether the object included in the analysis image is a dangerous object. The object determining module 313 receives the analysis image from the image analysis module 311 and determines whether the object included in the analysis image is a dangerous object that may cause a marine accident when the object collides with the ship. At this time, the object determining module 313 may further use the above-described visible distance adjusting module 250 to cross-check the determination result of the object.

That is, the object determination module 313 may further perform a process of comparing the dangerous object determined in the analysis image with the enlarged image through the visible-light adjustment module 250 to determine whether the object is a dangerous object.

The image storage module 315 stores the image. If there is no object in the analysis image analyzed by the image analysis module 311, the image storage module 315 may save the image for use in machine learning, for example. In addition, the image storage module 315 may further include a reference image. The reference image may be an image that can be used for calculating the distance between the dangerous object and the ship in the collision determination unit 133, which will be described later, and may be an image in which no object exists.

On the other hand, if the image processing unit 131 determines that there is a dangerous object, the collision determination unit 133 determines whether there is a risk of collision where the dangerous object may collide with the ship. For this, the collision determination unit 133 may include a distance measurement module 331 and a collision risk determination module 335.

The distance measurement module 331 calculates the distance between the dangerous object and the vessel by receiving the analysis image determined by the image processing unit 131 that a dangerous object exists. For this, the distance measurement module 331 can use the number of pixels and the pixel distance of the reference image and the analysis image previously stored in the image storage module 315.

Table 1 below shows the relationship between the number of pixels per section and the pixel distance. At this time, the number of pixels of the reference image and the analysis image is 2 million pixels.

Referring to Table 1, the distance measurement module 331 according to an embodiment of the present invention can use the actual distance and the pixel distance to calculate the distance between the ship and the dangerous object. That is, an image up to a measurement distance (20 m to 200 m) is prepared using a reference image, the number of pixels for each section is calculated, the number of pixels in an analysis image in which a dangerous object exists is counted, The distance between hazardous objects can be measured.

On the other hand, the distance between the ship and the hazardous object measured by the distance measurement module 331 is transmitted to the collision determination module 335. The collision determination module 335 determines the collision alarm type using the distance between the measured vessel and the dangerous object and the predetermined threshold distance value.

That is, when the distance between the ship and the hazardous object measured by the distance measurement module 331 is 100 m or more on the assumption that the predetermined threshold distance value is 100 m, the collision determination module 335 generates a warning alarm inside the ship Shipboard crew members may be warned that dangerous objects are present in front of them.

If the distance between the ship and the dangerous object measured by the distance measurement module 331 is less than 100 m, the collision determination module 335 may generate a danger warning inside the ship and warn that the collision risk is high.

In addition, the collision determination module 335 generates a danger warning on the outside of the ship, thereby notifying the user of the position of the ship on which the dangerous object is being operated so as to avoid the collision You may. For this, the collision determination module 335 may use sound or light as a danger warning to be outputted to the outside.

The collision determiner 133 may further include a velocity measurement module 333 according to an embodiment of the present invention. The speed measurement module 333 can measure the relative speed between the ship and the dangerous object. At this time, the speed measurement module 333 may use the information used in the distance measurement module 331 to measure the relative speed.

Through this, if the collision determination module 335 determines that the ship and the dangerous object move in different directions even if the ship approaches the ship using the relative speed, even if there is a dangerous object near the critical distance value, The determination module 335 may generate a warning signal inside the ship to prepare for a situation.

If the collision determination module 335 determines that the collision can not be avoided as a result of checking the path and distance between the dangerous object and the vessel using both the distance measurement module 331 and the velocity measurement module 333, An accident signal including the flight information of the ship in operation can be transmitted to the navy, the coast guard, or the paramedics through the communication network.

Through this, the navy, seafarers or paramedics can determine the condition of the ship before an accident occurs, and it can minimize human, material, and environmental damage through quicker response.

4 to 6 show an alarm method for preventing collision of a ship according to an embodiment of the present invention. Hereinafter, an alarm method for preventing collision of a ship according to an embodiment of the present invention will be described in detail with reference to FIG. 4 to FIG. 4 to 6, for the sake of convenience, an alarm system for preventing collision of a ship according to an embodiment of the present invention shown in FIGS. 1 to 3 is used. However, the present invention is not limited to this, It is apparent that the present invention can be carried out by using various devices, devices, and systems suitable for carrying out the present invention.

FIG. 4 is a flow chart briefly illustrating an alarm method for preventing collision of a ship using an alarm system for preventing collision of a ship according to an embodiment of the present invention, FIG. 5 is a flowchart showing a step of processing the image of FIG. FIG. 4 is a flow chart showing in more detail the step of anticipating the collision and the step of providing the notification.

Referring to FIG. 4, an alarm method 400 for preventing collision of a ship according to an embodiment of the present invention includes steps S410, S420, S430, And providing an alarm (S440).

First, an image is taken by an alarm system for preventing collision of a ship (step S410). The warning system for preventing collision of a ship photographs the direction of the ship in the outside of the ship and the direction of the ship is preferably the direction in which the image is taken, but the present invention is not limited thereto and may take various directions according to the user's needs .

Next, the ship collision warning system processes the image photographed in step S410 (step S420). The warning system for preventing collision with a ship can acquire an object that may collide with the ship by correcting and analyzing the captured image in step S410. To this end, step S420 may be formed to include steps S510 to S540 as shown in FIG.

The step S420 of processing the image of the ship collision avoidance alarm system may include a step S510 of correcting the mist, a step S520 of analyzing the image, a step S530 of adjusting the visible distance, (S540).

First, the ship collision warning system generates a corrected image by correcting the fog in the image photographed at step S410 for processing the image (step S510). At this time, if fog is not included in the shot image, the ship collision warning system may perform step S520 without generating a corrected image.

On the other hand, when the captured image includes fog, the warning system for preventing collision of the ship can generate the corrected image by removing the fog effect through the step of removing the fog (S511) and the step of performing the image correction (S513) have.

At this time, the warning system for preventing collision of a ship may be set to operate when it is determined that fog occurs when the visual distance decreases to a certain distance or less in order to remove fog in step S511, or the user may directly control on / Alternatively, the external image obtained in step S410 may be checked, and may be set to operate when the difference in brightness in the image hardly appears.

In addition, the ship collision warning system can use the MDCP algorithm named as Median Dark Channel Prior to remove fog from the shot image. Using this, the ship collision warning system can remove fog in a way that improves the contrast of the fog pixels in the RGB color space.

On the other hand, when the alarm system for preventing collision of the ship performs step S511 of removing fog by using the MDCP algorithm, the contrast of the mist part is forcibly improved, so that the distortion of color and the loss of edge value existing in the mist part may occur .

To compensate for this, the ship collision warning system can further correct for color distortion and edge value loss occurring after removing the fog in step S513. For this purpose, the accuracy of the color of the fog image in the HIS color space is measured An algorithm may be used to perform correction for color distortion and edge value loss.

Next, the ship collision warning system analyzes the corrected image or the shot image in which there is no fog to check whether there is any moving object (step S520). In this case, the moving object is not only an object having its own speed but also an object having a relative speed according to the operation of the ship.

The ship collision warning system analyzes the corrected image and confirms whether the object is captured in the image. The ship collision warning system can detect an object in a correction image and generate an analysis image using a machine learning and a Haar-like feature technique.

In other words, the warning system for ship collision avoidance can detect natural phenomena such as waves, waves, rain, snow, etc. from the front of ship by learning repetition of detection method using difference in brightness between area and area, And can detect a reef or other ship as an object in the image. In addition, the ship collision warning system can further increase the object detection capability by using the Adaboost algorithm if necessary.

If it is determined in step S520 that there is a moving object photographed in the image as a result of the image analysis in step S520, the alarm system for preventing collision of the ship performs step S530 of adjusting the visible distance. Through this, the ship collision warning system can further acquire an enlarged image for a certain period of time.

On the other hand, if it is determined in step S530 that an object exists in the image, the warning system for preventing collision with a ship may be an object having a danger that a moving object may collide with the ship, and further, It is further confirmed whether the object is a dangerous object (step S540).

The ship collision warning system determines whether the object included in the analysis image generated in step S520 is a dangerous object. The ship collision warning system receives the analysis image generated in step S520 and determines whether the object included in the analysis image is a dangerous object having a risk of causing a marine accident when the object collides with the ship. At this time, the ship collision warning system may further use the enlarged image of step S530 to cross-check the determination result of the object.

That is, the ship collision warning system can compare the images of step S520 and step S530 to determine whether the object included in the image is a dangerous object.

At this time, if it is determined in step S540 that the dangerous object is not included in the image, the ship collision warning system may perform step S550 to store the image for use in machine learning.

On the other hand, if it is determined in step S540 that the dangerous object is included in the image, the ship collision warning system can perform the step of estimating the collision risk (S430) and the step of providing the alarm according to the collision risk (S440) have.

A flowchart showing these steps S430 and S440 in more detail is shown in Fig. Referring to FIG. 6, in an alarm method for preventing collision of a ship according to an embodiment of the present invention, an alarm system for preventing a collision of a ship may include a step (S610) of calculating a distance to a dangerous object, And determining whether the dangerous object has entered a dangerous distance (S620).

The ship collision warning system calculates the distance between the dangerous object and the vessel (step S610) when it is determined in step S420 that the dangerous object exists in front of the ship.

The warning system for preventing collision with a ship can calculate the distance between the dangerous object and the ship by using the pre-stored reference image and the number of pixels and the pixel distance of the analysis image acquired in step S520.

The alarm system for preventing collision with a ship can calculate the distance between the dangerous object and the ship by using the relationship between the number of pixels per section and the pixel distance as shown in Table 1. Referring to Table 1, The alarm system prepares the image up to the measurement distance (20m ~ 200m) using the reference image, calculates the number of pixels for each section, counts the number of pixels in the analysis image where the dangerous object exists, The distance between the ship and the hazardous object can be measured.

On the other hand, the warning system for preventing collision with a ship determines whether a dangerous object has entered a dangerous distance by using the result of calculating the distance between the dangerous object and the ship in step S610 (step S620). In this case, the danger distance may be a critical distance value from a predetermined ship.

That is, if the predetermined danger distance is 100 m, if the distance between the ship and the dangerous object obtained in step S610 is 100 m or more, the internal alarm notification of step S640 may be performed. Through this, the warning system for prevention of collision with ship can alert the ship crew to dangerous objects ahead by generating a warning alert inside the ship.

If the distance between the ship and the dangerous object obtained in step S610 is less than 100m, which is less than the critical distance, the ship collision warning system may perform the external alarm notification in step S630. Through this, an alarm system for preventing collision of ships can generate a danger warning on the outside of the ship and allow the ship to recognize and respond to the position and approach of the ship when the dangerous object is in operation.

On the other hand, when performing the external alarm notification in step S630, the alarm system for preventing collision with the ship may continuously perform the internal alarm notification in step S640. This not only informs the location and access of the ship to the outside, but also alerts the crew that there is a dangerous object approaching the ship.

Meanwhile, step S430 of predicting the collision risk among the alarm methods for preventing collision of a ship according to an embodiment of the present invention includes calculating a distance to a dangerous object (S610), determining whether the dangerous object is in a dangerous distance And calculating a relative speed with the dangerous object between step S620 and step S615.

The ship collision warning system calculates the relative speed between the ship and the dangerous object (step S615). At this time, the ship collision warning system may use the distance information of step S610 to calculate the relative speed.

An alert system for preventing collision with a ship can determine whether a dangerous object approaching through a relative speed is a ship or a reef that is stationary, a moving marine creature or a ship, and may acquire the moving direction of a dangerous object and a ship.

The ship collision warning system uses the relative speed calculated in step S615 to determine whether a dangerous object exists at a position closer to the critical distance value when it is determined that the ship and the dangerous object move in different directions even if they approach each other, Even if a warning alert is generated inside the ship, it can be prepared for the situation.

If it is determined that the collision can not be avoided as a result of checking the path and distance between the dangerous object and the ship by further performing step S615, the collision avoidance alert system may transmit an accident signal Can be transmitted to the navy, sea port, or emergency service through the communication network.

Through this, the navy, seafarers or paramedics can determine the condition of the ship before an accident occurs, and it can minimize human, material, and environmental damage through quicker response.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

100: Ship collision prevention alarm system
110: photographing unit 130: ship collision forecasting unit
131: Image processor 133:
210: camera module 230: fog correction module
250: Visible distance adjustment module 311: Image analysis module
313: Object determination module 315: Image storage module
331: Distance measurement module 333: Speed measurement module
335: Collision risk determination module

Claims (27)

A photographing unit which photographs the outside of the ship and is formed so as to increase the visibility distance that can be identified through fog correction when fog is present; And
And a ship collision predicting unit for receiving an image acquired by the photographing unit and analyzing the image to confirm whether or not an object is photographed in the image, and determining a risk of collision with the object to generate an alarm,
Wherein,
A camera module for photographing the outside of the ship to acquire the image;
A fog is removed by using a median dark channel priority (MDCP) in an image acquired through the camera module, color distortion correction of the image and edge value correction of the image are further performed, A fog correction module for increasing the fog correction module; And
And a visible distance adjustment module for adjusting the magnification of the image using the variable zoom lens and enlarging the image for a predetermined time when the object is photographed on the image,
The ship collision prediction unit
An image processing unit for analyzing the image and checking whether the object is photographed in the image; And
And a collision judging unit for judging whether there is a collision risk that the ship may collide with the object identified by the image processing unit,
Wherein the image processing unit comprises:
An image analysis module that analyzes the image subjected to the fog correction and confirms whether the object is photographed on the image using a Haar-like feature technique;
An image storage module that stores the image and uses the stored image as a basic image in which the object is not captured when the image analysis module analyzes the object as not being photographed; And
And an object determining module for acquiring an image enlarged from the photographing unit to determine whether the object is a dangerous object that may possibly threaten the hull of the ship by colliding with the ship such as another ship or a reef, system. delete delete delete delete delete delete delete delete delete delete The method according to claim 1,
The collision determination unit may include a distance measurement module that calculates a distance to a dangerous object, which is an object with which the ship may collide; And
And a collision determination module for determining whether the dangerous object is in a collision danger distance by using the distance calculated by the distance measurement module and determining a collision alarm type. 13. The method of claim 12,
The distance measurement module measures the distance to the dangerous object by using the number of pixels and the pixel distance in the reference image photographed during the actual 10m unit period measured in the period, Alarm system. 13. The method of claim 12,
The distance measurement module may further include a speed measurement module for calculating a relative speed between the ship and the dangerous object. 13. The method of claim 12,
Wherein the collision determination module generates a warning when the distance from the dangerous object calculated by the distance measurement module is longer than the collision danger distance, and when the distance to the dangerous object is closer to the collision danger distance, To prevent collision of the ship. 16. The method of claim 15,
Wherein the collision determination unit outputs the warning alert to the inside of the ship and outputs the warning alert to the outside of the ship. A method for predicting a collision risk of a ship using an alarm system for preventing collision of a ship,
A fog removing unit that obtains an image by capturing an image of the outside of the ship and performs fog correction of the image using the median dark channel flier (MDCP) when the fog exists outside the vessel, ;
Generating a corrected image by performing color distortion correction of the fog removed image and edge value correction of the image;
Analyzing the corrected image to analyze whether the object has been photographed using the low-level feature technique, and, when the object is photographed, enlarging the corrected image for a predetermined period of time, Determining whether the object is a dangerous object that may collide with a ship and posing a threat to the ship, storing the corrected image if the object is not determined to be the dangerous object;
Calculating a distance between the dangerous object and the vessel; And
Determining whether the dangerous object is in a collision danger distance, and generating an alarm. delete delete delete delete delete 18. The method of claim 17,
Calculating the distance includes calculating the distance between the dangerous object and the ship by using the number of pixels and the pixel distance in the reference image photographed per actual 10m unit period measured in the past. 24. The method of claim 23,
Wherein the reference image is photographed at 2 megapixels. 24. The method of claim 23,
Wherein the step of calculating the distance further comprises calculating a relative speed of the dangerous object with respect to the ship. 18. The method of claim 17,
The generating of the alert may include generating a warning alert inside the ship if the dangerous object is not in the collision danger distance and if the dangerous object is in the collision danger distance, More alarming method for preventing collision of ship. 27. The method of claim 26,
The alarm generating method further includes the step of transmitting accident information including at least one of the current position, the speed and the time of the ship through the communication network when the ship is unable to avoid the dangerous object.

Images