Abstract:
PBFUC (Predicting Bus Fitness Using CNN) may play a great role to reduce traffic
accidents, classification of bus condition. Giving safe roads to people and overcoming from
the rate of accidents due to fitness of bus, for all the reasons Bus Fitness Prediction system
is essential. We traced that issue of road accidents and settled us to work for reducing road
accidents. The most eminent model like Convolutional Neural Network (CNN) is good for
classification. In this paper, we proposed a model of Convolutional Neural Network (CNN)
for classifying fitness of buses. To classify the condition or fitness of buses we followed
two parts – one is feature extraction and another one is classification. These two parts
performed good via CNN in our system. In this system, we worked on 2 classes to classify
bus fitness- one class is fit buses and another one is unfit buses. So, CNN model played a
great role to find a good accuracy and build our system perfectly. We split our total data as
80% for training dataset and 20% for testing dataset there after we achieved evaluate
accuracy 80.93% for our CNN model and 84.56% for training accuracy.
