KR102297147B1 - Three-dimensional Image Analysis Method for Evaluation of Visual Function in Fish - Google Patents

Abstract

The three-dimensional image analysis method for evaluating the visual function of fish according to an embodiment of the present invention includes the steps of removing a background image excluding a plurality of fish from a first image and a second image obtained by a three-dimensional stereo camera, respectively ; performing object matching on the first image and the second image from which the background image is removed; generating a disparity map between the first image and the second image; generating a 3D image by combining the first image and the second image with reference to the disparity map; and obtaining 3D location information of the plurality of fish from the 3D image.

Description

{Three-dimensional Image Analysis Method for Evaluation of Visual Function in Fish}

The present application relates to a three-dimensional image analysis method for evaluating the visual function of fish.

Research using fish such as zebrafish for toxicity screening of various chemical substances has been widely conducted.

For example, in order to screen for ocular toxicity of chemical substances, research is being conducted on a method of providing a predetermined visual stimulus to a zebrafish and analyzing the movement of the zebrafish in response thereto.

Conventionally, in order to analyze the movement of the zebrafish, a two-dimensional image of a zebrafish responding to a visual stimulus was photographed, and the movement of the zebrafish was analyzed by observing the photographed two-dimensional image with the naked eye.

However, a lot of time and effort is required for image analysis, and furthermore, there are cases in which it is difficult to accurately identify a zebrafish in a two-dimensional image, so that the accuracy of the analysis result may be deteriorated.

Therefore, in the art, in the toxicity screening of chemical substances using fish such as zebrafish, there is a need for a method for more quickly and accurately analyzing the movement of fish.

In order to solve the above problems, an embodiment of the present invention provides a three-dimensional image analysis method for evaluating the visual function of fish.

The three-dimensional image analysis method for evaluating the visual function of the fish includes: removing a background image excluding a plurality of fish from each of the first image and the second image acquired by a three-dimensional stereo camera; performing object matching on the first image and the second image from which the background image is removed; generating a disparity map between the first image and the second image; generating a 3D image by combining the first image and the second image with reference to the disparity map; and obtaining 3D location information of the plurality of fish from the 3D image.

Incidentally, the means for solving the above problems do not enumerate all the features of the present invention. Various features of the present invention and its advantages and effects may be understood in more detail with reference to the following specific embodiments.

According to an embodiment of the present invention, in the toxicity screening of chemical substances using fish such as zebrafish, the movement of the fish is faster and faster by acquiring a three-dimensional image of the fish and analyzing it to obtain three-dimensional position information. can be accurately analyzed.

1 is a schematic diagram of an apparatus for evaluating visual function of fish to which the present invention can be applied.
FIG. 2 is a diagram showing an implementation example of the apparatus for evaluating the visual function of fish shown in FIG. 1 .
3 is a flowchart of a 3D image analysis method for evaluating the visual function of a fish according to an embodiment of the present invention.

Hereinafter, preferred embodiments will be described in detail so that those of ordinary skill in the art can easily practice the present invention with reference to the accompanying drawings. However, in describing a preferred embodiment of the present invention in detail, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. In addition, the same reference numerals are used throughout the drawings for parts having similar functions and functions.

In addition, throughout the specification, when a part is 'connected' with another part, it is not only 'directly connected' but also 'indirectly connected' with another element interposed therebetween. include In addition, 'including' a certain component means that other components may be further included, rather than excluding other components, unless otherwise stated.

1 is a schematic diagram of an apparatus for evaluating visual function of fish to which the present invention can be applied, and FIG. 2 is a diagram illustrating an embodiment of the apparatus for evaluating visual function of fish shown in FIG. 1 .

1 and 2 , the apparatus 100 for evaluating the visual function of a fish to which the present invention can be applied includes a moving unit 110 , a display unit 120 , a three-dimensional image acquisition unit 130 , and an image analysis unit ( 140) may be included.

The moving unit 110 may be configured to include at least one lane through which the fish can move.

According to an embodiment, the moving unit 110 is made of a transparent material, and accordingly, when a moving image is displayed on the display unit 120 to be described later, the fish may visually respond to the moving image and move along the lane.

The display unit 120 is provided to be detachably attached to the lower portion of the moving unit 110 , and a moving picture having a preset pattern of stimulation during water flow may be displayed.

Here, the water flow visual stimulus pattern refers to a moving picture pattern that shows the water flow and at the same time has a visual stimulus pattern of a repeating pattern of a certain color.

By rheotaxis, we mean the movement of cells or animals in response to the stimulus of flowing water.

In general, when water flows, the flow of light occurs in the opposite direction to the flow of water. In the case of larvae, they tend to move along the stream of light.

The present invention introduces the mainstream of these fishes into the visual function evaluation. Specifically, when a visual stimulus pattern video having a repeating pattern of a certain light (color) is played, the visual stimulus pattern video of the repeating pattern is recognized as a flow of light in the opposite direction to the water flow, and fish with normal vision respond to this. This will move in the direction of the visual stimulus of the visual stimulus pattern video.

In the present invention, an image of a moving fish can be obtained by displaying a moving image having a stimulus pattern during a water flow with a difference in contrast sensitivity for each stage, that is, a difference in color contrast sensitivity.

According to one embodiment, as shown in the table below, based on the color difference of the color table divided into 16 stages of G1 to G16, one or more water flow stimulation patterns having a difference in sensitivity for each stage may be used.

[graph]

As described above, the display unit 120 can be used with any device that can display a moving picture having a stimulus pattern during a water flow having a difference in contrast sensitivity. For example, any portable display implementation device, such as a smartphone or tablet, may be used.

The 3D image acquisition unit 130 is installed to be spaced apart from the upper side of the moving unit 110 by a predetermined distance, and visually responds to a video displayed through the display unit 120 to move within the lane of the moving unit 110 . A three-dimensional image of a plurality of fish may be acquired.

According to an embodiment, the 3D image acquisition unit 130 may be implemented as a 3D stereo camera. A 3D stereo camera acquires a 3D image using two cameras. It is classified into a parallel type and an orthogonal type according to the camera arrangement method, and can be classified into a horizontal type, a cross type, and a horizontal movement type according to the gaze point control method. have.

Although FIG. 2 illustrates an example in which the 3D image acquisition unit 130 is implemented using an orthogonal stereo camera, the present invention is not necessarily limited thereto. That is, the 3D image acquisition unit 130 may be implemented by various well-known stereo cameras or stereoscopic cameras.

The image analysis unit 140 may receive and analyze the 3D image acquired by the 3D image acquisition unit 130 to obtain 3D position information of a plurality of fish. The image analysis unit 140 may acquire 3D position information of a plurality of fish according to the 3D image analysis method for evaluating the visual function of the fish shown in FIG. 3 , which will be described in detail later with reference to FIG. 3 . do it with

Also, the image analyzer 140 may count the number of fish according to the three-dimensional position information of the plurality of fish, and may analyze the movement pattern of the plurality of fish based on the counted number of fish.

Also, the image analyzer 140 may statisticize the distribution in the height direction of the plurality of fish based on the 3D position information.

The image analyzer 140 may be implemented in various computing devices such as, for example, a desktop, a notebook computer, and a smart phone.

As described above, the image analysis unit 140 obtains the 3D position information of the fish by analyzing the 3D image, so that the 3D position of each of the plurality of fish moving in the moving unit 110 can be accurately identified. , and based on this, it is possible to accurately count the number of fish, and furthermore, it is possible to improve the analysis accuracy of the movement pattern of the fish.

In addition, in the analysis of the movement pattern of the fish based on the distribution in the height direction of the fish statisticalized by the image analysis unit 140 , an object showing an unwanted movement pattern (eg, vertical movement) is excluded from the analysis target, or , it can be expected to further activate research using fish such as zebrafish by utilizing the distribution in the height direction of fish for toxicity screening analysis.

3 is a flowchart of a 3D image analysis method for evaluating a fish's visual function according to an embodiment of the present invention.

Referring to FIG. 3 , first, a first image and a second image acquired by a 3D stereo camera may be input ( S310 ). Here, the first image and the second image may be a left image and a right image obtained simultaneously from the left and right sides of the moving part in which the plurality of fish move.

Thereafter, the background image may be removed from each of the input first image and the second image ( S320 ). Here, the background image may be an image of a stimulus during water flow provided from the lower part of the fish (ie, the bottom surface of the moving part through which the fish moves).

Thereafter, if necessary, the first image and the second image from which the background image is removed may be divided into preset sizes, that is, the first image and the second image may be divided into a plurality of images having sizes suitable for image processing, respectively ( S330).

Thereafter, object matching may be performed on the first image and the second image divided into a plurality of images ( S340 ). Here, object matching refers to matching the same object in the first image and the second image acquired at the same time. First, the object matching is performed by identifying the same object (ie, fish) in both images and calculating the distance between the objects. can do.

Thereafter, a disparity map between the first image and the second image may be generated ( S350 ). Here, the disparity map is information indicating a difference between corresponding points in the first image and the second image, that is, the position of the same object, and is a disparity map using the distance and pixel distance between the object and the 3D stereo camera. can create

Thereafter, if necessary, noise except for a plurality of fish may be removed by an active filter ( S360 ). Here, the active filter may selectively remove unnecessary noise except for a plurality of fish by actively adjusting the amount of noise filtering according to the state of the input image.

Thereafter, a 3D image may be generated by combining the first image and the second image with reference to the disparity map ( S370 ).

Thereafter, 3D position information of the object may be obtained from the 3D image (S380). A method known to those skilled in the art may be applied to a method of obtaining 3D position information of an object from a 3D image, and a detailed description thereof will be omitted.

The present invention is not limited by the above embodiments and the accompanying drawings. For those of ordinary skill in the art to which the present invention pertains, it will be apparent that the components according to the present invention can be substituted, modified and changed without departing from the technical spirit of the present invention.

100: fish visual function evaluation device
110: moving unit
120: display unit
130: 3D image acquisition unit
140: image analysis unit

Claims (4)

removing a background image excluding a plurality of fish from each of the first image and the second image obtained by the three-dimensional stereo camera;
performing object matching on the first image and the second image from which the background image is removed;
generating a disparity map between the first image and the second image;
generating a 3D image by combining the first image and the second image with reference to the disparity map; and
Comprising the step of obtaining three-dimensional position information of the plurality of fish from the 3D image,
A three-dimensional image for the evaluation of visual function of fish, characterized in that the movement pattern of the fish is analyzed based on the distribution in the height direction of the fish statistically based on the three-dimensional position information of the fish and used for the evaluation of the visual function of the fish analysis method.
The method of claim 1,
The first image and the second image are a three-dimensional image analysis method for evaluating the visual function of a fish, characterized in that it is a left image and a right image acquired simultaneously from the left and right sides, respectively, for the moving part in which the plurality of fish move.
The method of claim 1,
The three-dimensional image analysis method for evaluating the visual function of a fish, characterized in that it further comprises the step of dividing the first image and the second image from which the background image is removed into preset sizes.
The method of claim 1,
The 3D image analysis method for evaluating the visual function of a fish, characterized in that it further comprises the step of removing noise except for a plurality of fish by an active filter on the first image and the second image from which the background image is removed.

Images