# US 2022/0036053 A1 · Method and apparatus for identifying animal species

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0036053 A1
- Document ID: US20220036053A1
- Status: Published application
- Kind: A1
- Application number: 17/279,661
- Filed: 2019-04-18
- Published: 2022-02-03
- Earliest claimed: 2018-10-16
- Assignees: Korea University Research and Business Foundation
- Applicants: Korea University Research and Business Foundation
- Inventors: Hanseok KO; Sangwook PARK; Kyung-Deuk KO; Donghyeon KIM
- Web page: https://patentgrep.com/patent/US20220036053A1

## CPC

- G06V 40/10: Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06F 18/254: {of classification results, e.g. of results related to same input data}
- G06N 3/02: Neural networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G06V 10/811: {the classifiers operating on different input data, e.g. multi-modal recognition}
- G06V 10/993: {Evaluation of the quality of the acquired pattern}
- G10L 17/10: Multimodal systems, i.e. based on the integration of multiple recognition engines or fusion of expert systems
- G10L 17/26: Recognition of special voice characteristics, e.g. for use in lie detectors; Recognition of animal voices
- G06N 3/045: Combinations of networks
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Disclosed are a method and an apparatus for identifying animal species by using audiovisual information. A method for identifying animal species, according to one embodiment of the present invention, may include: a step of receiving an input signal for an object to be identified; a step of processing image information and acoustic information based on the input signal, wherein a processing result of the image information and a processing result of the acoustic information are represented by class-specific scores; a step of determining whether the image information processing result and the acoustic information processing result corresponding to the input signal exist; and a final result derivation step of fusing the image information processing result and the acoustic information processing result according to the determination result and classifying the object to be identified as a certain animal species by using the fused processing result.

## Claims

### Claim 1

A method for identifying animal species comprising:
a step of receiving an input signal for an object to be identified;
a step of processing image information and acoustic information based on the input signal, wherein a processing result of the image information and a processing result of the acoustic information are represented by class-specific scores;
a step of determining whether the image information processing result and the acoustic information processing result corresponding to the input signal exist; and
a final result derivation step of fusing the image information processing result and the acoustic information processing result according to the determination result, and classifying the object to be identified as a certain animal species by using the fused processing result.

### Claim 2 (depends on 1)

The method for identifying the animal species of claim 1, wherein the final result derivation step includes a step of fusing the score of the image information processing result and the score of the acoustic information processing result when both the image information processing result and the acoustic information processing result exist.

### Claim 3 (depends on 1)

The method for identifying the animal species of claim 1, wherein the final result derivation step includes a step of classifying the object to be identified as a certain animal species by using any one of the scores of the image information processing result and the score of the acoustic information processing result when only one of the image information processing result and the acoustic information processing result exists.

### Claim 4 (depends on 2)

The method for identifying the animal species of claim 2, wherein the final result derivation step derives the final result by using Equation 1 below.
finalresult
=
arg
⁢
⁢
max
c
⁢
(
∑
n
=
1
N
⁢
w
c
n
⁢
s
~
c
n
)
[
Equation
⁢
⁢
1
]
(c and n represent indexes for an object to be recognized and a system, respectively, wcn means the fidelity of a c-th object to be recognized in an nth system, scn means a score for the c-th object to be recognized in the nth system, and scn means a score normalized by dividing scn by
∑
c
=
1
C
⁢
S
c
n
.
)

### Claim 5 (depends on 4)

The method for identifying the animal species of claim 4, wherein wcn in Equation 1 above is calculated by Equation 2 below.
w
c
n
=
P
⁡
(
G
=
c
❘
O
=
c
)
=
P
⁡
(
G
=
c
,
O
=
c
)
∑
g
⁢
P
⁡
(
G
=
g
,
O
=
c
)
[
Equation
⁢
⁢
2
]
(wherein, G means an actual value and O means a recognition result obtained from the recognition system.)

### Claim 6 (depends on 6)

The method for identifying the animal species of claim 6, wherein the image information and acoustic information processing step includes a step of processing the image information and the acoustic information by using a convolutional neural network (CNN) technique.
