This study presents pre-processing methods for detecting lane detection using camera and Light Detection and Ranging (LiDAR) sensor technologies. Standard image processing methods are not suitable for complicated roads with various sign on the ground. Thus, determining the right techniques for pre-processing such data would be a challenge. The objectives of this study are to pre-process the scanned images and apply the image recognition algorithm for lane detection. The study employed Canny Edge Detection and Hough Transform algorithms on several sets of images. A different region of interest was experimented to find the optimal one. The experimental results showed that the proposed algorithms could be practical in terms of effectively detecting road lines and generate lane detection.
Feature selection has been widely applied in many areas such as classification of spam emails, cancer cells, fraudulent claims, credit risk, text categorisation and DNA microarray analysis. Classification involves building predictive models to predict the target variable based on several input variables (features). This study compares filter and wrapper feature selection methods to maximise the classifier accuracy. The logistic regression was used as a classifier while the performance of the feature selection methods was based on the classification accuracy, Akaike information criteria (AIC), Bayesian information criteria (BIC), Area Under Receiver operator curve (AUC), as well as sensitivity and specificity of the classifier. The simulation study involves generating data for continuous features and one binary dependent variable for different sample sizes. The filter methods used are correlation based feature selection and information gain, while the wrapper methods are sequential forward and sequential backward elimination. The simulation was carried out using R, an open-source programming language. Simulation results showed that the wrapper method (sequential forward selection and sequential backward elimination) methods were better than the filter method in selecting the correct features.