Method and electronic device for extracting appearance information of animal

Abstract

The objective of an embodiment of the present invention is to provide a method and an electronic device capable of appropriately extracting appearance information of an animal even from an image including an object that interferes with identity recognition of the animal. A method for extracting appearance information of an animal according to an embodiment of the present invention comprises the steps of: obtaining an input image including a body of an animal; segmenting an animal body region and a non-interest region from the input image; and extracting fur color information by processing the non-interest region as excluded from the animal body region.

Description

Method for extracting animal external appearance information and electronic device

[0001] The present invention relates to a method for extracting external appearance information of an animal and an electronic device.

[0002] In modern society, there is a rising demand for companion animals that can live alongside humans and provide emotional support. Along with the increase in the population living with companion animals, the problem of lost (or abandoned) animals is also on the rise. To address this issue, animal registration systems are being implemented to register animal identity information, and technologies utilizing information and communication technology to identify animals are being introduced.

[0003] To identify animal identity, technologies that recognize biometric information through video recording are being introduced as a replacement for existing external or internal recognition chips. For example, technologies are being introduced that identify animals by using the patterns of wrinkles on a dog's nose (nasal prints) or a cat's facial features as biometric data. Such image-based biometric recognition is garnering significant attention for its user convenience, as identification can be easily performed via smartphones.

[0004] For example, if a passerby with a smartphone discovers a lost animal and films the animal, the animal's appearance information can be identified from the filmed video, and identity information (e.g., name, age, owner information) can be identified from the animal's appearance information.

[0005] However, unlike humans, animals have a habit of not recognizing the act of being filmed and continuing to move, and furthermore, they may run away if a stranger films them, so it is difficult to obtain proper external information even when filming animals. In particular, when extracting fur color information of a dog, there is a problem in that it is difficult to extract accurate fur color information even when filming if the dog sticks out its tongue or if there are interfering objects such as leashes.

[0006] Embodiments of the present invention aim to provide a method and an electronic device capable of appropriately extracting external appearance information of an animal even from an image containing an object that interferes with the identification of the animal.

[0007] A method for extracting external appearance information of an animal according to an embodiment of the present invention comprises: acquiring an input image including the body of an animal; dividing the animal body region and a non-interest region in the input image; and excluding the non-interest region from the animal body region to extract fur color information.

[0008] In an embodiment of the present invention, the non-interest area may include an area corresponding to a collar or tongue.

[0009] In an embodiment of the present invention, the step of dividing the animal body region and the non-interest region may include: a step of dividing the background region and the animal body region in the input image to extract the animal body region; and a step of extracting the non-interest region from the animal body region.

[0010] In an embodiment of the present invention, the step of extracting fur color information may include: a step of extracting a color distribution in the remaining area excluding the non-interest area in the animal body area; and a step of clustering colors having a similarity within a certain range in the color distribution to extract main color information of the remaining area.

[0011] In an embodiment of the present invention, the main color information may include RGB values ​​of main colors extracted through color clustering and ratio values ​​occupied by each of the main colors in the remaining area.

[0012] An electronic device for extracting external appearance information of an animal according to the present invention includes a memory; and a processor electrically connected to the memory. The processor is configured to acquire an input image including the body of an animal through the camera, divide the animal body region and a non-interest region in the input image, and extract fur color information by excluding the non-interest region from the animal body region.

[0013] According to an embodiment of the present invention, by dividing an animal body region and a non-interest region in an input image and excluding the non-interest region from the animal body region to extract fur color information, external appearance information of an animal can be effectively extracted even from an image containing objects that interfere with the identification of the animal.

[0014] Figure 1 illustrates the reporting and discovery process when a lost animal occurs.

[0015] FIG. 2 illustrates an electronic device according to the present invention.

[0016] FIG. 3 is a flowchart illustrating a method for extracting external appearance information of an animal according to the present invention.

[0017] Figure 4 illustrates the process of extracting color information as external appearance information of an animal.

[0018] Figure 5 is a diagram illustrating the process of extracting color information according to a conventional method.

[0019] FIG. 6 is a diagram illustrating the process of extracting color information according to the present invention.

[0020] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present invention. The present invention may be embodied in various different forms and is not limited to the embodiments described herein.

[0021] To clearly explain the present invention, parts unrelated to the explanation have been omitted, and the same reference numerals are used for identical or similar components throughout the specification.

[0022] In addition, in various embodiments, components having the same configuration are described using the same reference numerals only in the representative embodiment, and in other embodiments, only configurations different from the representative embodiment are described.

[0023] Throughout the specification, when a part is described as being "connected (or combined)" with another part, this includes not only cases where they are "directly connected (or combined)" but also cases where they are "indirectly connected (or combined)" with other members in between. Furthermore, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0024] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0025] Hereinafter, a method and electronic device for extracting external appearance information of an animal according to the present invention will be described. FIG. 1 illustrates a process for reporting and finding a lost animal when it occurs. The lost animal reporting and finding service according to the present invention can be performed by a server (20) and a plurality of user devices (10A, 10B) connected to the server (20). The lost animal reporting and finding service can be executed by the user's device (10A) registering the animal's identity information and / or biometric information to the server (20). Alternatively, even if the user's device (10A) has not registered the animal in advance, the lost animal reporting and finding may be possible even if the user registers the animal's identity information and a video of the animal taken in advance with the server (20) after the animal is lost.

[0026] The process of finding, reporting, and extracting external appearance information of a lost animal according to the present invention can be executed by an application installed on each user device (10A, 10B) and by a server (20) that provides services through the application. Here, the application includes applications that operate on mobile devices such as smartphones or PCs (Personal Computers) and applications that run on a web basis.

[0027] The user device (10A, 10B) is an electronic device that the user carries and uses. The user device (10A, 10B) may be a portable electronic device such as a smartphone, a tablet device, or a laptop computer. Additionally, the user device (10A, 10B) may be a wearable device such as a smart watch or smart glass. The user device (10A, 10B) may be an electronic device such as a PC (personal computer) or a game console. The user device (10A, 10B) may be an electronic device equipped with a processor, memory, a display, and a communication module capable of running an application program for biometric recognition of a pet.

[0028] When a user device (10A) that has lost an animal reports the lost animal to a server (20) (S10), the server (20) can register the reported animal as a lost animal by including identity information and biometric information such as appearance information. The server (20) can provide information about the lost animal to various user devices, government offices, or animal protection organizations through an application.

[0029] Meanwhile, if a lost animal is found by another user, the user device (10B) can take a picture of the animal and transmit it to the server (20) (S20). Here, the user device (10B) may transmit the original captured animal photo to the server (20), or extract the animal's appearance information from the user device (10B) and transmit it to the server (20).

[0030] The server (20) evaluates whether the image of the animal received from the user device (10B) matches the animal reported as lost, and sends a message notifying the owner of the animal found to the user device (10A) of the animal determined to match (S30). The user device (10A) checks through the message received from the server (20) whether it is its own animal, and if it is its own animal, it can find the animal through the server (20) or directly contact the user device (10B) that found the animal to find the animal.

[0031] In the present invention, animal fur color information may be used as external appearance information for animal identification. The process of extracting animal fur color information can be performed by extracting body parts of the animal from an input image in which the animal is captured and using the color distribution of the body parts. However, since the animal's body parts in the image may include other parts in addition to fur, there is a risk that the fur color information may be distorted. For example, in the case of a dog, the tongue may be included in the body parts as it often sticks out its tongue, and if it is wearing a leash, the leash may be included in the body parts.

[0032] The present invention provides a method for extracting external appearance information of an animal and an electronic device for performing the same, which can recognize the fur color of an animal even when an object that interferes with the recognition of the animal's fur color is included during the imaging of the animal.

[0033] FIG. 2 illustrates the structure of an electronic device (100) according to the present invention. The electronic device (100) according to the present invention may correspond to one of the user device (10A, 10B) or the server (20) of FIG. 1. In this document, the server (20) receives an image of an animal from the user device (10B) and extracts external appearance information of the animal from the image as an example, but the scope of the present invention is not limited thereto and includes identifying external appearance information of the animal at the user device (10B).

[0034] An electronic device (100) according to the present invention includes a memory (150) and a processor (110) electrically connected to the memory (150). Additionally, the electronic device (100) may include a communication module (120) for transmitting or receiving data with another device, a display (130) for visually outputting information, and an input / output interface (140) for connecting with another device or for inputting and outputting information.

[0035] The processor (110) can perform operations and data processing for the operation of the electronic device (100), and control each module of the electronic device (100). The processor (110) can execute one or more software programs stored in memory (150). The processor (110) may include a general-purpose processor (e.g., CPU) that runs an operating system managing the overall operation of the electronic device (100). Additionally, the processor (110) may include a dedicated processor (e.g., GPU, NPU) for specific computation processing. The processor (110) may include one or more processing circuits. The method of operation of the electronic device (100) described below may be executed by the processor (110).

[0036] The communication module (120) provides an environment in which the electronic device (100) can communicate with an external entity (e.g., an authentication server, a base station, an IoT device, a user device). The communication module (120) may include a device for performing wired or wireless communication. The communication module (120) may provide data received from an external entity to the processor (110). The communication module (120) may convert the data provided from the processor (110) and transmit it to the external entity.

[0037] The display (130) is a device for outputting visual information and may include a panel and a control circuit. The memory (150) may store programs, such as an operating system and an application for the operation of the electronic device (100), and data used during the execution of the program. The memory (150) may include volatile memory (e.g., DRAM) and non-volatile memory (e.g., flash memory). Additionally, the memory (150) may include a mass storage device for organizing a database.

[0038] FIG. 3 is a flowchart illustrating a method for extracting external appearance information of an animal according to the present invention. The process of extracting external appearance information of an animal described below can be executed by a processor (110) of an electronic device (100).

[0039] A method for extracting external appearance information of an animal according to the present invention comprises the steps of: acquiring an input image including the body of an animal (S310); dividing the animal body region and a non-interest region in the input image (S320); and excluding the non-interest region from the animal body region to extract fur color information (S330).

[0040] In step S310, the processor (110) acquires an input image captured by a camera. The processor (110) may receive an image from an external device, and the processor (110) may acquire an input image including the body of an animal through a camera provided in the electronic device (100).

[0041] In step S320, the processor (110) can divide the animal body region and the non-interest region in the input image. That is, the processor (110) can extract the animal body region corresponding to the animal's body in the input image and extract the non-interest region to be excluded from the animal body region. The non-interest region refers to the remaining area other than the area containing the fur color to be extracted. For example, the non-interest region includes areas corresponding to a collar, tongue, teeth, clothing, and jewelry (accessories). The non-interest region includes all areas that interfere with recognizing the animal's fur color.

[0042] In the present invention, the step of dividing the animal body region and the non-interest region (S320) includes the step of dividing the background region and the animal body region in the input image to extract the animal body region, and the step of extracting the non-interest region from the animal body region. The processor (110) can primarily extract the animal body region from the background image in the input image, and secondarily extract the non-interest region in which at least a portion overlaps the animal body region.

[0043] In step S330, the processor (110) excludes non-interest regions from the animal body region and extracts fur color information for the remaining region. The processor (110) can extract fur color information for the remaining region excluding non-interest regions from the animal body region of the input image.

[0044] The step of extracting fur color information (S330) may include extracting a color distribution in the remaining area excluding the non-interest area in the animal body region, and extracting main color information of the remaining area by clustering colors having similarity within a certain range in the color distribution. The processor (110) may identify a color distribution for the remaining area excluding the non-interest area in the animal body region, cluster colors that can be treated as substantially the same color in the color distribution into one main color, and extract main color information as a result of color clustering. As a clustering method, a method of referencing a predefined look-up table or a K-means clustering method may be used. The main color information may include the RGB values ​​of the main colors extracted through color clustering and the ratio values ​​that each main color occupies in the remaining area.

[0045] FIG. 4 illustrates the process of extracting color information as external shape information of an animal. Referring to FIG. 4, when an input image is input to an electronic device (100), the area where the animal is located, i.e., object coordinate information, is extracted from the input image through a segmentation model. The segmentation model is an algorithm that outputs location information occupied by an object from the input image. The segmentation model can be implemented by a deep neural network. The object coordinate information is the coordinates that constitute the area occupied by the animal's body part in the input image. When the animal body area corresponding to the object coordinate information is extracted from the input image, the processor (110) checks the color distribution (RGB value distribution) in the animal body area and performs color clustering (RGB) on the animal body area. Color clustering is the process of grouping pixels with similar RGB values ​​within a certain range for each pixel's RGB component value in the animal body area and treating them as a single main RGB value. Main color information can be output through color clustering.

[0046] Here, the processor (110) can perform color clustering on the remaining areas excluding non-interest areas in the animal body region. FIG. 5 is a diagram illustrating the process of extracting color information according to a conventional method, and FIG. 6 is a diagram illustrating the process of extracting color information according to the present invention.

[0047] Referring to FIG. 5, an input image (IMG) containing an animal is acquired as in (a), and an animal body region (R1) is extracted from the input image (IMG) as in (b). As in (c), the background region is removed from the input image (IMG) and color clustering is performed on the animal body region (R1). As a result of color clustering, the RGB values ​​of three main colors and the ratio of each main RGB value in the animal body region (R1) are output as main color information as in (d). At this time, since the dog's tongue is included in the animal body region (R1), it may interfere with recognizing the animal's fur color.

[0048] Referring to FIG. 6, an input image (IMG) containing an animal is acquired as in (a), and an animal body region (R1) is extracted from the input image (IMG) as in (b), where the non-interest region (N1) corresponding to the dog's tongue is removed from the animal body region (R1). Through color clustering of the remaining region from which the non-interest region (N1) has been removed from the animal body region (R1) as in (c), three main RGB values ​​and the ratio of each main RGB value in the remaining region are output as main color information as in (d).

[0049] In this way, by detecting non-interest regions that interfere with the process of extracting animal fur color information from an input image and extracting color information from the remaining area after removing the non-interest regions from the animal body region, accurate animal fur color information can be obtained even when objects that interfere with the extraction of animal appearance information are captured together.

[0050] The embodiments and drawings attached to this specification merely clearly illustrate a part of the technical concept included in the present invention, and it is obvious that variations and specific embodiments that can be easily deduced by a person skilled in the art within the scope of the technical concept included in the specification and drawings of the present invention are all included within the scope of the rights of the present invention.

[0051] Accordingly, the scope of the present invention should not be limited to the described embodiments, and all things equivalent to or having equivalent variations to the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the concept of the present invention.

Claims

1. In a method for extracting external appearance information of animals, A step of acquiring an input image including the body of an animal; A step of segmenting animal body regions and non-interest regions in the input image above; and A method comprising the step of extracting fur color information by excluding the non-interest area from the animal body area.

2. In Paragraph 1, The above non-interest area is a method that includes an area corresponding to a collar or tongue.

3. In Paragraph 1, The step of dividing the above animal body region and non-interest region is, A step of extracting the animal body region by dividing the background region and the animal body region in the input image above; and A method comprising the step of extracting the non-interest region from the animal body region.

4. In Paragraph 1, The step of extracting the above hair color information is, A step of extracting a color distribution in the remaining area excluding the non-interest area in the animal body area; and A method comprising the step of clustering colors having similarity within a certain range in the above color distribution to extract main color information of the remaining area.

5. In Paragraph 4, The above main color information includes RGB values ​​of main colors extracted through color clustering and ratio values ​​of each main color occupying the remaining area.

6. In an electronic device for extracting external appearance information of an animal, Memory; and It includes a processor electrically connected to the above memory, and The above processor is, Acquire an input image including the animal's body, and Segment the animal body region and non-interest region in the above input image, and An electronic device configured to extract fur color information by excluding the non-interest area from the above animal body region.

7. In Paragraph 6, The above non-interest area is an electronic device including an area corresponding to a collar or tongue.

8. In Paragraph 6, The above processor is, The background region and the animal's body region are separated from the input image to extract the animal's body region, and An electronic device configured to extract the non-interest region from the above animal body region.

9. In Paragraph 6, The above processor is, Extract the color distribution in the remaining area excluding the non-interest area from the above animal body area, and An electronic device configured to extract main color information of the remaining area by clustering colors having similarity within a certain range in the above color distribution.

10. In Paragraph 9, The above main color information is an electronic device comprising RGB values ​​of main colors extracted through color clustering and ratio values ​​occupied by each of the main colors in the remaining area.

Images