Traffic Vision, Tracking and Counting Using Machine Learning and Machine Vision
Başlık:
Traffic Vision, Tracking and Counting Using Machine Learning and Machine Vision
Yazar:
Wang, Haiyan, author.
ISBN:
9780438069947
Yazar Ek Girişi:
Fiziksel Tanımlama:
1 electronic resource (38 pages)
Genel Not:
Source: Masters Abstracts International, Volume: 57-06M(E).
Advisors: Mohammad Pourhomayoun Committee members: Russell Abbott; Raj Pamula.
Özet:
The goal of this research is to design and develop a system based on advanced machine learning and machine vision to detect, monitor, and count pedestrians and cyclists in video streams captured by traffic cameras. This research is in collaborations with the City of Los Angeles and Toyota Foundation. The output results of the developed system will be used to help city planners develop better and safer travel environments and make our urban transportation safer for our people, especially for pedestrians and bicyclists.
In this research, we have developed an end-to-end system including a series of computer vision algorithms, various machine learning algorithms including deep neural networks, and optimal state estimator algorithms that receive video streams in real-time, and monitor, recognize, track, and count pedestrians and cyclists.
Our video data is real-life data captured from 5 different traffic cameras, each camera has different angles and faces a different area, contains footages of pedestrians and cyclists. In this study, we encountered with many challenging conditions while processing the video streams such as different lightening conditions, including very strong or dark lighting, shadows, different weather conditions, pedestrians and cyclists may walk or ride in groups or be hidden when passing a wall or a street sign for certain times. Therefore, in our project, we needed to develop and use multiple algorithms to handle many different challenges.
Here, we will describe and compare different methods we used for our research, as well as show our approach for solving challenging cases in different video data.
Notlar:
School code: 0962
Konu Başlığı:
Tüzel Kişi Ek Girişi:
Mevcut:*
Yer Numarası | Demirbaş Numarası | Shelf Location | Lokasyon / Statüsü / İade Tarihi |
---|---|---|---|
XX(692607.1) | 692607-1001 | Proquest E-Tez Koleksiyonu | Arıyor... |
On Order
Liste seç
Bunu varsayılan liste yap.
Öğeler başarıyla eklendi
Öğeler eklenirken hata oldu. Lütfen tekrar deneyiniz.
:
Select An Item
Data usage warning: You will receive one text message for each title you selected.
Standard text messaging rates apply.