SISTEM PENGATUR TRAFFIC LIGHT MENGGUNAKAN JARINGAN SARAF TIRUAN BACKPROPAGATION BERDASARKAN VOLUME KENDARAAN
Total View This Week24
Total View This Week24
Institusion
Universitas Muhammadiyah Malang
Author
Subject
TK Electrical engineering. Electronics Nuclear engineering
Datestamp
2019-08-01 08:54:47
Abstract :
Congestion is a problem for every city anywhere. Various solutions are pursued by adding roads, expanding roads and so on. Traffic Light is an effort to reduce congestion. But the existing Traffic Light works by setting a predetermined duration of time for each segment. One solution to reduce the risk of congestion, it is necessary to design a Traffic Light system that can work more effectively and flexibly, where the setting of the duration of the green light must depend on the density of the passing vehicle. Then there will be different Traffic Light settings in each segment, according to the density of passing vehicles. The research will be carried out in the form of a system for managing traffic lights, which is proposed in the form of a system with the detection of vehicle objects and how many vehicles are contained in each lane. Object detection is done using the haar cascade method. In recognizing an object and for the introduction of an object type using Backpropagation Artificial Neural Network. Haar Cascade will ignore image areas that do not have objects that meet the criteria. This is very helpful in improving the performance of the classifier. The effect, the possibility of object detection errors is reduced and the detection accuracy is increased.
Congestion is a problem for every city anywhere. Various solutions are pursued by adding roads, expanding roads and so on. Traffic Light is an effort to reduce congestion. But the existing Traffic Light works by setting a predetermined duration of time for each segment. One solution to reduce the risk of congestion, it is necessary to design a Traffic Light system that can work more effectively and flexibly, where the setting of the duration of the green light must depend on the density of the passing vehicle. Then there will be different Traffic Light settings in each segment, according to the density of passing vehicles. The research will be carried out in the form of a system for managing traffic lights, which is proposed in the form of a system with the detection of vehicle objects and how many vehicles are contained in each lane. Object detection is done using the haar cascade method. In recognizing an object and for the introduction of an object type using Backpropagation Artificial Neural Network. Haar Cascade will ignore image areas that do not have objects that meet the criteria. This is very helpful in improving the performance of the classifier. The effect, the possibility of object detection errors is reduced and the detection accuracy is increased.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
