KR20160049569A - System and method for providing plant information using smart device - Google Patents

Abstract

The present invention relates to a system and a method for providing plant information using a smart device, which includes: a smart device for providing photographing guide information regarding a target plant, determining whether the target plant exists in a photographed image, and requesting for the plant information of the target plant by using the photographed image; and a server for extracting the plant information of the target plant, with respect to the photographed image transmitted from the smart device, from a plant information database through a plant extract algorithm to transmit the plant information to the smart device. According to the present invention, an accurate image with respect to a corresponding plant is obtained by photographing with the smart device depending on the photographing guide, more accurate image is secured by performing a target determination which primarily determines whether the corresponding plant exists based on the image photographed by the smart device, and accurate plant information is provided by analyzing the corresponding plant through the plant extract algorithm at the server.

Description

TECHNICAL FIELD [0001] The present invention relates to a system and a method for providing plant information using a smart device,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system and a method for providing plant information using a smart device, and more particularly, to a system and method for providing plant information using a smart device, And more particularly, to a system and method for providing plant information using a device.

There have been many proposals for providing plant information using smart devices.

As one method, there is a technique of confirming plant information by requesting and downloading plant information through communication between a smart device and a server. However, this technique has a disadvantage in that it requires a large amount of downloaded data to be directly retrieved.

In order to compensate for these drawbacks, another method is to transmit a plant image to a server after the plant is photographed in a smart device, to retrieve the plant image, and to download and output plant information from the server. However, since this technique is performed only on close-up photographing in a smart device and transmitted to the server, that is, since the photographing guide is not set, there is a problem that errors occur in analyzing the plant in the server.

Korean Patent Publication No. 10-2011-0127473 (published on November 25, 2011)

[1] "Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images" ECSE-626: Statistical Computer Vision (ICCV 2001) [2] "A Method for Regional Feature Extraction of Flower Images," International Conference on ICIP-2010: Yong Pei, and Weiqun Cao (2010) [3] "Tutorial: Image Segmentation", Graduate Institute of Communication Engineering National Taiwan University, Taipei, Taiwan, ROC: Yu-Hsiang Wang

SUMMARY OF THE INVENTION The present invention has been made in order to solve the problems of the prior art described above. Accordingly, it is an object of the present invention to provide a smart device that enables shooting through a photographing guide, And to provide plant information providing system and method using a smart device that enables accurate plant information to be provided through a plant extracting algorithm in a server.

According to another aspect of the present invention, there is provided a system for providing plant information using a smart device, the system comprising: providing photographing guide information about a target plant; determining whether the target plant exists in the photographed image; A smart device for requesting plant information of the plant; And a server for extracting the plant information of the target plant from the plant information database through the plant extraction algorithm for the photographed image transmitted from the smart device and transmitting the plant information to the smart device.

In this case, the smart device preferably includes a discriminator implemented by performing machine learning on the plant to confirm whether the smart device is photographed according to the classification system of the plant selected in the photographing guide.

In addition, the server preferably includes a plant information database for analyzing the plant, extracting features, and comparing the extracted plant information through the plant extraction algorithm.

Meanwhile, a plant information providing method using a smart device of the present invention includes the steps of: (a) providing a photographing guide in a smart device so that an attribute of a target plant to be photographed can be selected; (b) checking whether the smart device is photographed in accordance with the classification scheme of the target plant selected through the photographing guide in response to the photographing of the target plant; And (c) separating the target plant and the background using a CROP algorithm in the server, and performing a search in the plant information database.

In the step (a), the photographing guide preferably provides a photographing frame of the target plant.

The step (b) is preferably performed through a discriminator realized by performing machine learning on a plant.

The crop algorithm can selectively use a method of dividing the color difference between the center area and the edge area, filling a part corresponding to the background with black, and a method of cutting the image according to the size of the shooting frame in the smart device .

As described above, according to the plant information providing system and method using the smart device according to the present invention, it is possible to acquire an accurate image of the plant by photographing in accordance with the photographing guide in the smart device.

In addition, according to the present invention, a more accurate image can be obtained by performing object discrimination for discriminating the presence or absence of the plant from the image photographed by the smart device.

According to the present invention, accurate plant information can be provided by analyzing the plant through the plant extraction algorithm in the server.

1 is a conceptual diagram illustrating an operation of a plant information providing system using a smart device according to an embodiment of the present invention.
2 is a configuration diagram of a plant information providing system using a smart device according to an embodiment of the present invention.
3 is a flowchart of a plant information providing method using a smart device according to an embodiment of the present invention.
4 is a flowchart illustrating an object determination process according to an exemplary embodiment of the present invention.
5 is a flowchart of a plant extraction algorithm according to an embodiment of the present invention.

Hereinafter, a plant information providing system and method using the smart device of the present invention will be described in detail with reference to the accompanying drawings.

1 is a conceptual diagram illustrating an operation of a plant information providing system using a smart device according to an embodiment of the present invention.

Referring to FIG. 1, first, a photographing guide corresponding to a property to be photographed (for example, a flower shape, a number of petals) is input to a smart device Output. The user confirms the photographing guide and photographs the target plant. In the smart device, it is determined whether the target plant exists in the photographed image (object discrimination). At this time, the object discrimination may be performed through its own analysis in the smart device, or the captured image may be transmitted to the server to receive the analysis result. On the other hand, if it is not possible to obtain the analysis results as a result of analyzing the smart device itself at the time of object discrimination, it is also preferable to receive the analysis result through the server. Here, the object discrimination may include not only determining the presence or absence of a plant in the photographed image, but also extracting plant information as detailed information on the plant.

Then, when the object discrimination is completed, the photographed image is transmitted to the server. In the server, the plant is analyzed by the plant extracting algorithm to extract the features, the extracted plant information is compared with the plant information database, .

As described above, the present invention minimizes the error through the photographing guide, object discrimination, and plant extraction algorithm, and provides more accurate plant information with higher recognition rate.

2 is a configuration diagram of a plant information providing system using a smart device according to an embodiment of the present invention.

Referring to FIG. 2, the plant information providing system using the smart device of the present invention provides photographing guide information on a target plant through a UI (User Interface) 11, A smart device 1 for discriminating the presence or absence of a plant and requesting plant information of the target plant by using the photographed image through the communication unit 13 and a photographed image transmitted from the smart device 1 via the communication unit 21. [ And a server 2 for extracting the plant information of the target plant from the plant information database 23 through execution of the plant extraction algorithm in the image analysis unit 22 and transmitting the plant information to the smart device 1.

Here, in providing the photographing guide, for example, the smart device 1 compresses the flower shape into five types based on the classification system for the flower shape registered in the "National Biology Information System" , The number of petals receives user input for less than 10, and if it is more than that, it is provided as "unknown". Further, it is preferable to provide a photographing frame of a circular shape at the time of photographing so as to obtain a flower shape of a certain size.

In addition, the smart device 1 confirms whether it is photographed according to the classification system of the plant (for example, flower) selected in the photographing guide through the discriminator implemented by machine learning on the plant using the machine learning program provided by OPenCV It is preferable to output the processed image through the video processing section 14. [

Here, the process in which the discriminator learns flowers is briefly described.

The photographed image of the flower is scanned over the whole photographed image by using an infrared sensor, and the sensor data is obtained by scanning. The photographing frame provided by the photographing guide is divided into a set number of unit cells. A representative value representative of each unit cell is calculated using sensor data measured in each divided unit cell. And extracts a feature vector of a flower having a representative value for each calculated unit cell as a factor. When the feature vector is extracted, the extracted feature vector factors are compared to learn the type of the flower. If the factor of the feature vector is too large, it is difficult to distinguish the type of the flower. Therefore, desirable. This is to reduce unnecessary parts of the feature vector to reduce the object to be computed and to extract only the representative feature vectors that are important in authenticity discrimination. Accordingly, it is preferable to reduce the dimension of the extracted feature vector by applying the linear feature extraction method, and to select only a representative representative feature vector representing the feature of the flower.

The server 2 crops a part of the plant (for example, flower part) through a plant extraction algorithm in the image analysis part 22 so that the captured image received from the smart device 1 can be compared. And then extracts HSV color values. Here, the plant extraction algorithm uses two CROP algorithms. First, the method of dividing the color difference between the center area and the corner area and filling the background part with black color is preferentially used. In this CROP process, A method of cutting out the image according to the size of the photographing frame photographed by the smart device 1 is used. Then, HSV color value is extracted by averaging the color of the flower leaf part and flower center part. On the other hand, if the HSV color value is not extracted from the server 2, the candidate list may be provided. Here, HSV (Hue Saturation Value) means a method of assigning a color to a hue, a saturation, and a brightness as RGB (Red, Green, Blue).

If the image analysis target is a flower, the flower is distributed at a specific position in the HSV space according to the color system. In the image analysis unit 22, The color can be distinguished.

Hereinafter, a plant information providing method using the smart device of the present invention using the system configured as described above will be described.

3 is a flowchart of a plant information providing method using a smart device according to an embodiment of the present invention.

S (Step) 1: The user selects an attribute (for example, flower shape, number of petals) for the target plant to be photographed through the photographing guide. In choosing attributes, for example, the shape of flowers is provided in five types based on the classification system for flower shapes registered in the "National Biology Information System", and the number of petals is 10 And receives the user input for the following, and if it is more than that, it is provided as "unknown ".

S2: The target plant is then photographed in consideration of the photographed frame.

S3: Using the machine learning program provided by OPENCV, machine learning for the plant is carried out, and it is confirmed by the discriminator implemented that it is photographed according to the plant classification system selected in the photographing guide. That is, an object discrimination for discriminating the presence or absence of the target plant in the photographed image is performed. At this time, it is preferable to transmit the processed image in the object discrimination process.

S4: When the object discrimination is completed, the smart device 1 transmits the photographed image to the server 2.

S5: The server 2 extracts the HSV color value after CROP of a part of the plant (for example, flower part) through a plant extraction algorithm.

S6: Next, the server 2 searches in the plant information database 23 based on the extracted data (the average of the flower leaf parts and the flower center parts). At this time, in the search, the flower shape (data value provided from the user through the photographing guide), the number of petals (the data range (1 to 10, unknown) provided by the user through the photographing guide) The plant information database 23 searches for an object on the basis of color (range of average color values of each region) and flowering time (value based on photographing time).

S7: When the search is completed, the search result is returned to the smart device 1.

S8: The smart device 1 outputs the representative image and the flower name.

S9: The user is asked to confirm that the search result matches the target plant.

S10: Provide detailed information (flower name, flower language, habitat, flowering time, etc.) of the flower when the search result matches the target plant. On the other hand, if the search result does not match the target plant, a candidate list can be provided.

4 is a flowchart illustrating an object determination process according to an exemplary embodiment of the present invention.

Referring to FIG. 4, when the target plant is photographed, it is determined whether the target plant exists in the image through the discriminator realized by machine learning on the flower using the machine learning program (S21 to S22).

If the target plant exists, the photographed image is transmitted to the server 2 (S23). Otherwise, the resume message is output (S24).

5 is a flowchart of a plant extraction algorithm according to an embodiment of the present invention.

Referring to FIG. 5, first, a part of a plant (for example, flower part) is cropped (S51).

The color difference between the center area and the corner area is distinguished, and the part corresponding to the background is filled with black. In step S52, the target plant is separated by deleting a portion having a value other than the foreground, after assigning a value to the coordinates according to the color difference based on the clear foreground and background.

If the background and the target plant are not separated, that is, if a problem occurs in the CROP process, it is determined whether or not the background and the target plant are separated from each other according to the size of the shooting frame photographed by the smart device 1 in 'Step 1' (S53 to S54). This separates the plant from the background.

On the other hand, when the background and the target plant are separated, HSV color values are extracted by averaging the colors of the flower leaf part and flower central part (S55).

As described above, the present invention can provide more accurate plant information by more accurately discriminating plants and extracting more accurate plant information through a photographing guide, an object discrimination, and a plant extraction algorithm. That is, it is possible to provide the photographing guide to the user so that accurate information can be input, and more precise information can be provided at a higher recognition rate through object discrimination before transmitting the image to the server 2, In the analysis step for separating the necessary parts, different algorithms can be applied to minimize errors. As described above, the plant information observed through the server 2 providing various plant information is acquired as an image and transmitted to the server 2, thereby analyzing the corresponding plant image and easily and accurately There is an advantage to be provided. In addition, detailed information about plants that the user needs can be provided from the server 2 using the smart device 1, so that information about the plant can be easily grasped, and there is an advantage that it is educationally advantageous.

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, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the invention.

1: Smart devices
14:
2: Server
22: Image analysis section

Claims (7)

A smart device that provides photographing guide information on a target plant, determines whether or not the target plant exists in a photographed image, and requests plant information of the target plant using the photographed image; And
And a server for extracting the plant information of the target plant from the plant information database through the plant extraction algorithm for the photographed image transmitted from the smart device and transmitting the plant information to the smart device.
The method according to claim 1,
Wherein the smart device comprises a discriminator implemented by performing a machine learning on a plant in order to confirm whether the smart device is photographed according to the classification system of the plant selected in the photographing guide.
The method according to claim 1,
Wherein the server includes a plant information database for analyzing the plant to extract features and extracting the extracted plant information through the plant extraction algorithm.
(a) providing, in a smart device, a photographing guide so that an attribute of a target plant to be photographed can be selected;
(b) checking whether the smart device is photographed in accordance with the classification scheme of the target plant selected through the photographing guide in response to the photographing of the target plant; And
(c) separating the target plant and the background using a CROP algorithm in the server, and performing a search in a plant information database.
5. The method of claim 4,
Wherein the photographing guide in the step (a) provides a photographing frame of the target plant.
5. The method of claim 4,
Wherein the step (b) is performed through a discriminator implemented by machine learning on a plant.
5. The method of claim 4,
Wherein the crop algorithm comprises:
A method of dividing a color difference between a center area and an edge area and filling a part corresponding to a background with black and a method of cutting a frame according to the size of a frame in the smart device, Delivery method.

Images