Final Project Signal Digital Processing Science Data - Universitas Negeri Surabaya
Автор: Al-FirdausUnesa
Загружено: 2025-12-19
Просмотров: 8
Описание:
Rizki Piji Fathoni - 24031554029
Abdullah Al-Firdaus Nuzula - 24031554008
FINAL PROJECT DIGITAL SIGNAL PROCESSING
DESIGN OF A VEHICLE SOUND CLASSIFICATION SYSTEM BASED ON ADAPTIVE FILTERING AND MULTI-PARAMETER SPECTRAL FEATURE EXTRACTION
The objective of this research is to use digital signal processing techniques to create an autonomous vehicle classification system based on auditory inputs. Sound recordings of a variety of vehicles, including buses, trains, trucks, cars, motorcycles, bikes, airplanes, and helicopters are included in the dataset. Mel-frequency cepstral coefficients (MFCC), spectral centroid, spectral rolloff, zero-crossing rate (ZCR), and spectral analysis using the fast Fourier transform (FFT) and signal preprocessing with a bandpass digital filter is different for each class to remove noise are all part of the applied methodology. The Random Forest machine learning algorithm was used for classification. The system's 98.52% accuracy in identifying vehicle kinds was demonstrated by test results, demonstrating the value of combining signal processing and feature extraction methods.
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
![Как сжимаются изображения? [46 МБ ↘↘ 4,07 МБ] JPEG в деталях](https://image.4k-video.ru/id-video/Kv1Hiv3ox8I)
