KR102207737B1 - An unmanned insert monitoring system and method of the insert recognition using the same - Google Patents

Abstract

The present invention relates to an unmanned organism monitoring system and a method for recognizing organisms using the same, and by going through the steps of binarizing, filtering, and analyzing a captured organism image for monitoring, the specific type and/or the number of organisms can be easily determined. It has the advantage of being able to classify and analyze.

Description

An unmanned insert monitoring system and method of the insert recognition using the same}

The present invention relates to an unmanned organism monitoring system and a method for recognizing organisms using the same.

The global environment has been constantly changing over a long period of time, and the ecosystem has been constantly responding and adapting to these changes. Starting in the 19th century, as the use of fossil fuels increased, the concentration of carbon dioxide in the atmosphere showed a sharp increase. As a result, global warming is causing not only local and regional climates, but also global climate change. The rapid climate change and the resulting environmental change within a short period of time are affecting not only population fluctuations or extinctions at the species level, but also population and community changes, and even the entire ecosystem. In particular, in the case of many species, environmental changes due to climate change caused by human activities are more significant for survival as well as maintenance of the species in a situation where the population is rapidly decreasing due to the artificial loss or fragmentation of habitats due to development and excessive use. It is a threat. These effects are taking place on a global scale and across ecosystems, and have great ripple effects in the business and economic sectors of the country and local communities. Therefore, for the maintenance of the ecosystem and further survival of humanity, it is necessary to understand and predict the changes in the ecosystem and to prepare countermeasures for the social impact of the changes in the ecosystem.

On the other hand, in the related art, a technology for aquatic organism monitoring system and method that can grasp and predict changes and impacts of an aquatic ecosystem due to climate change has been developed. The monitoring system is a device that attracts aquatic organisms to photograph with a photographing means according to a set time, stores the photographed image, and then measures the types of insects and the number of insects in the image to analyze aquatic organisms.

On the other hand, when analyzing aquatic organisms using images captured by the monitoring system, there is an inconvenience of having to classify the type and number of insects by directly analyzing the images by humans.

Therefore, there is a need to develop a program that can easily classify the type and number of insects using images of organisms such as aquatic insects.

Korean Patent Registration No. 10-1546453

In order to solve such a problem, the present invention is to provide a program capable of easily classifying the type and number of organisms using images of organisms photographed by a monitoring system.

To achieve the above object,

The present invention in one embodiment,

In the unmanned organism monitoring system comprising an attracting means for attracting a living creature, a photographing means for photographing the attracted creature, and an analysis means for analyzing the photographed image,

The analysis means includes an image conversion unit for converting the photographed color image into a black and white image;

A binarization unit for binarizing the black and white image into a binarized image;

A filter unit for classifying living things and non-living things by removing noise included in the binarized image; And

An analysis unit that compares the data of living organisms with pre-stored reference data in the image from which noise has been removed, and derives the type and number of organisms in the image; It provides an unmanned organism monitoring system comprising a.

In addition, the present invention in another embodiment,

Converting the color image photographed by the creature into a black and white image;

Binarizing the black and white image into a binarized image;

Classifying living things and non-living things by removing noise included in the binarized image; And

Comparing the data of living organisms and pre-stored reference data in the image from which noise has been removed, and deriving the type and number of organisms in the image; It provides a biological recognition method using the unmanned biological monitoring system comprising a.

According to the unmanned organism monitoring system and the method for recognizing a organism using the same according to the present invention, there is an advantage in that it is possible to easily classify and analyze the specific type and/or the number of the captured organism.

1 is a conceptual diagram schematically showing the configuration of an unmanned organism monitoring system according to an embodiment of the present invention.
2 is a flowchart of a method for recognizing a creature using an unmanned organism monitoring system according to an embodiment of the present invention.
3 is a photograph sequentially showing steps of analyzing an organism according to a method for recognizing a organism according to an embodiment of the present invention.

In the present invention, various modifications may be made and various embodiments may be provided, and specific embodiments will be illustrated in the drawings and described in detail in the detailed description.

However, this is not intended to limit the present invention to a specific embodiment, it is to be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention.

In the present invention, terms such as "comprises" or "have" are intended to designate the presence of features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, but one or more other features. It is to be understood that the presence or addition of elements or numbers, steps, actions, components, parts, or combinations thereof, does not preclude in advance.

In addition, the accompanying drawings in the present invention should be understood as being enlarged or reduced for convenience of description.

1 is a conceptual diagram schematically showing the configuration of an unmanned organism monitoring system according to an embodiment of the present invention, and FIG. 2 is a flow chart of a method for recognizing an organism using an unmanned organism monitoring system according to an embodiment of the present invention, and FIG. 3 Is a photograph sequentially showing steps of analyzing a creature according to the method for recognizing a creature according to an embodiment of the present invention.

Hereinafter, with reference to FIGS. 1 to 3, an unmanned organism monitoring system of the present invention and a method for recognizing a organism using the same will be described in detail.

The present invention in one embodiment,

In the unmanned organism monitoring system 100 comprising an attracting means 110 for attracting a living creature, a photographing means 120 for photographing an attracted creature, and an analysis means 130 for analyzing the captured image,

The analysis means 130 may include an image conversion unit 131 for converting the photographed color image into a black and white image;

A binarization unit 132 for binarizing a black and white image into a binarized image;

A filter unit 133 for classifying living things and non-living things by removing noise included in the binarized image; And

An analysis unit 134 that compares the data of the organism from the noise-removed image with the previously stored reference data to derive the type and number of organisms in the image; It provides an unmanned organism monitoring system 100 characterized in that it comprises a (see Fig. 1).

In the present invention, "living organisms" may mean all organisms that can be photographed using a monitoring system, may mean soil roaming invertebrates, daylight insects, or aquatic organisms, and may be aquatic organisms. Aquatic organisms refer to organisms living in the sea, rivers, lakes, or coasts thereof, and may be, for example, whirligigfish, water striders, jangguaebi, squirrels, stingrays, dragonflies, and fish.

As described above, the unmanned organism monitoring system according to an embodiment of the present invention includes an incentive means 110, a photographing means 120, and an analysis means 130.

First, the attracting means 110 may include a housing 111 that is a space that attracts living things, and an attracting device 112 that attracts living things to the enclosure 111. In order to monitor aquatic organisms, the enclosure 111 may submerge only a part of the enclosure 111 due to its own buoyancy, and may be suspended on a waterway such as a river. In addition, in the vicinity of or inside the enclosure 111, a attracting device 112 for attracting a creature to be monitored may be included. For example, it may be a light emitting diode (LED), a mercury lamp, a black light, or a manned lamp such as a halogen lamp, and may be a light emitting diode (LED).

A photographing means 120 for photographing the interior of the enclosure 111 may be included in the interior or top of the enclosure 111.

Specifically, the photographing means 120 includes a photographing unit 121, a photographing time adjustment unit 122 and a storage unit 123. The photographing unit 121 is a device for photographing underwater under the enclosure 111, and may be a commonly used digital camera or a video camera, for example, a high resolution camera of 12 million pixels as a digital camera. .

In addition, the photographing time adjustment unit 122 is configured to set the operation time of the photographing unit 121, and in the present invention, it may be set to shoot once per 10 minutes for about 17 hours from 6:00 to 23:00. Meanwhile, the storage unit 123 is a means for storing an image photographed by the photographing unit 121 and may be, for example, an SD card.

In the present invention, the analysis means 130 is a means for analyzing an image captured by a creature to be monitored, and includes an image conversion unit 131, a binarization unit 132, a filter unit 133, and an analysis unit 134. Include.

The image conversion unit 131 converts the color image received from the storage unit 123 of the photographing means 120 into a black and white image. This can implement a function of distinguishing an object from a background in the image through the process of segmentation of living things. Meanwhile, this may affect the recognition rate of the software depending on the sharpness of the original image.

The binarization unit 132 may convert the monochrome image converted by the image conversion unit 131 into a binarized image. Binarization is the expression of color values distributed in various ways for a pixel with only two values of 0 and 1. Specifically, the binarization unit 132 may set a threshold value based on an average gray level value or a standard deviation value of gray levels of a black and white image of the image. In addition, when the gray level of the pixel is less than the threshold value, the pixel is matched to 1, and when the gray level of the pixel is higher than the threshold value, the captured image is binarized by matching the pixel to 0.

The threshold may be appropriately set between 0 and 1. For example, a value of 0.4 can be set as a threshold, and if the gray level of the pixel is less than the threshold, the pixel is matched to 1, and if the gray level of the pixel is higher than the threshold, the pixel is matched to 0. Can be binarized. The part represented by 1 may be represented by white, and the part represented by 0 may be represented by black.

For example, a living entity may be represented as white, and a non-living entity may be represented as black.

In addition, the filter unit 133 may classify living things and non-living things by removing noise included in the binarized image. Specifically, after setting a threshold value based on an image pixel average value or a standard deviation value in the binarized image, it is classified as living and non-living based on the threshold value, and the non-living can be removed. Small objects that exist in the binarized image can be removed. For example, you can remove objects less than 10 pixels from the binarized image.

In one embodiment of the present invention, since the aquatic insect to be analyzed usually has a certain size, an individual having a size relatively smaller than this may be regarded as a foreign matter. Through this process, foreign matter existing in the background of the image is removed, resulting in a clear and smooth image.

Next, the analysis unit 134 may derive the type and number of organisms in the image by comparing the data of the organism from which the noise has been removed and the previously stored reference data. Specifically, the analysis unit 134 may derive the type and number of organisms in the image by comparing the pixels of the creature with the pixels of the reference data in the image from which noise has been removed.

First, you can set the target object to be analyzed in this process. Although there are various biological objects in the captured image, since the target object can have a certain size, only the target object can be expressed by removing excessively large or small objects. For example, when the target object is water polka dots, a radius can be obtained from the center point of the object in the image and a pixel can be measured according to the value of each area, which can be distinguished by comparing with a previously stored pixel value. Meanwhile, the ripples can be set to 21 pixels, the polka dots can be set to 14 pixels, and the water polka dots can be set to 4 pixels.

In addition, the analysis unit 134 may compare the pixel value with the pixel value in the image to derive the type and number of organisms to be analyzed.

In addition, the present invention in another embodiment,

Converting the color image photographed by the creature into a black and white image;

Binarizing the black and white image into a binarized image;

Classifying living things and non-living things by removing noise included in the binarized image; And

Comparing the data of living organisms and pre-stored reference data in the image from which noise has been removed, and deriving the type and number of organisms in the image; It provides a biological recognition method using the unmanned biological monitoring system 100 including (see FIGS. 2 and 3).

Specifically, according to the method for recognizing organisms according to the present invention, there is an advantage of being able to easily classify and analyze specific types and/or the number of organisms photographed using the unmanned organism monitoring system 100.

First, the photographing unit 121 of the unmanned creature monitoring system 100 installed at the location to be photographed takes a biological image at a preset time interval of the photographing time control unit 122, and stores the photographed image according to the present invention. It is stored in the unit 123. On the other hand, the operation time may be set to be taken once every 10 to 30 minutes from 5 to 6 o'clock to 22 to 23 o'clock by the photographing time adjusting means.

Next, a target organism may be recognized and analyzed using the captured image image. A color image photographed by a living organism may be converted into a black and white image, and the converted image may be subjected to binary processing in two colors. On the other hand, "binarization" refers to expressing color values that are variously distributed for a pixel with only two values of 0 and 1.

More specifically, the binarization process includes setting a threshold value based on an average gray level value or a standard deviation value of the gray level of the black and white image, and if the gray level of the pixel of the black and white image is less than the threshold value, the pixel is Matching and, if the gray level of the pixel is higher than the threshold value, binarizing the captured image by matching the pixel to 0.

The threshold may be appropriately set between 0 and 1. For example, a value of 0.4 can be set as a threshold, and if the gray level of the pixel is less than the threshold, the pixel is matched to 1, and if the gray level of the pixel is higher than the threshold, the pixel is matched to 0. Can be binarized. The part represented by 1 may be represented by white, and the part represented by 0 may be represented by black.

For example, a living entity may be represented as white, and a non-living entity may be represented as black.

Biological and non-living can be classified by removing noise included in the binarized image.

Specifically, it includes setting a threshold value based on the average pixel value or standard deviation value of the image in the binarized image, and classifying it into living things and non-living things based on the threshold value, and removing the non-living things. . Here, a small object existing in the binarized image can be removed. For example, you can remove objects less than 10 pixels from the binarized image.

In one embodiment of the present invention, since the aquatic insect to be analyzed usually has a certain size, an individual having a size relatively smaller than this may be regarded as a foreign matter. Through this process, foreign matter existing in the background of the image is removed, resulting in a clear and smooth image.

Finally, the method of recognizing a creature of the present invention includes the step of deriving the type and number of organisms in the image by comparing the data of the organism with the reference data previously stored in the image from which noise has been removed.

Specifically, the analysis unit 134 may derive the type and number of organisms in the image by comparing the pixels of the creature with the pixels of the reference data in the image from which noise has been removed.

First, you can set the target object to be analyzed in this process. Although there are various biological objects in the captured image, since the target object can have a certain size, only the target object can be expressed by removing excessively large or small objects. For example, when the target object is water polka dots, a radius can be obtained from the center point of the object in the image and a pixel can be measured according to the value of each area, which can be distinguished by comparing with a previously stored pixel value. On the other hand, the ripples can be set to 21 pixels, the polka dots can be set to 14 pixels, and the water polka dots can be set to 4 pixels.

In addition, the analysis unit 134 may compare the pixel value with the pixel value in the image to derive the type and number of organisms to be analyzed.

Accordingly, it is possible to easily monitor ecological changes such as the frequency and time of appearance of organisms according to climate change.

100: unmanned organism monitoring system
110: incentives
111: enclosure 112: attraction device
120: shooting means
121: photographing unit 122: photographing time adjustment unit
123: storage
130: analysis means
131: image conversion unit 132: binarization unit
133: filter unit 134: analysis unit

Claims (9)

In the unmanned organism monitoring system comprising an attracting means comprising a space that is a space for attracting a living thing and an attracting device that attracts a living thing to the enclosure, a photographing means for photographing the attracted creature, and an analysis means for analyzing the captured image,
The analysis means includes: an image conversion unit for converting a photographed color image into a black and white image in order to realize a function of classifying an object from a background in an image through a segmentation process of a living thing;
A binarization unit for binarizing the monochrome image into a binarized image by matching the pixel to 1 when the gray level of the pixel of the monochrome image is less than the threshold value and matching the pixel to 0 when the gray level of the pixel is higher than the threshold value;
A filter unit configured to set a threshold value based on an average image pixel value or a standard deviation value in the binarized image, and then classify it into living things and non-living things based on the threshold value, and removing the non-living things; And
An analysis unit that compares a pixel of a living being from an image from which noise has been removed and a pixel of a pre-stored reference data to derive the type and number of living things in the image; Including
The threshold value in the filter unit is 10 pixels, and the filter unit classifies and removes objects less than 10 pixels as non-living,
The attracting device is made of any one of an LED, a mercury lamp, a halogen lamp, and a black light,
The unmanned organism monitoring system, characterized in that the organism is a daylight insect.
delete delete delete The method of claim 1,
The photographing means includes a photographing unit for photographing an attracted creature, a photographing time adjustment unit for setting an operation time of the photographing means, and a storage means for storing an image photographed by the photographing unit.


delete delete delete delete

Images