ACOUSTIC EVENT DETECTION METHOD IN DEEP LEARNING-BASED DETECTION ENVIRONMENT

• Main Authors: PARK, In Young, KIM, Hong Kook
• Format: Patent
• Language: English; French; Korean
• Published: 03.06.2021
• Subjects: ACOUSTICS; ALARM SYSTEMS; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTROPHONIC MUSICAL INSTRUMENTS; HANDLING RECORD CARRIERS; MUSICAL INSTRUMENTS; ORDER TELEGRAPHS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SIGNALLING; SIGNALLING OR CALLING SYSTEMS

Disclosed is a method which: extracts multi-features from acoustic data of interest from acoustic data containing one or more sound sources; classifies the one or more sound sources included in the multi-features using an artificial intelligence model based on fast region-convolutional neural network-long short-term memory (Fast Region-CNN-LSTM); and detects an event using the classified sound sources. La présente invention concerne un procédé : qui extrait de multiples caractéristiques à partir de données acoustiques dignes d'intérêt à partir de données acoustiques contenant une ou plusieurs sources sonores ; qui classe la ou les sources sonores incluses dans les multiples caractéristiques à l'aide d'un modèle d'intelligence artificielle basé sur une mémoire à court et long terme de réseau neuronal à région rapide (Fast region-CNN-LSTM) ; et qui détecte un événement à l'aide des sources sonores classées. 본 개시는 적어도 하나 이상의 음원이 포함된 음향 데이터에 기초하여 사익 음향 데이터에 포함된 복합특징값을 추출하고, 고속 지역 합성곱 장단기 기억 신경망(Fast Region-CNN-LSTM)을 기반으로 하는 인공 지능 모델을 이용하여 상기 복합특징값에 포함된 적어도 하나 이상의 음원을 분류하고, 분류된 음원을 이용하여 사건을 감지하는 방법을 개시한다.