Nursabillilah Mohd Ali

Performance Comparison between RGB and HSV Color Segmentations for Road Signs Detection

This paper compares the performance of RGB and HSV color segmentations method in road signs detection. The road signs images are taken under various illumination changes, partial occlusion and rotational changes. The proposed algorithms using both RGB and HSV color space are able to detect the 3 standard types of colored images namely Red, Yellow and Blue. The experiment shows that the HSV color algorithm achieved better detection accuracy compared to RGB color space.

Traffic Sign Detection and Classification for Driver Assistant System

In this paper we explain the proposed method of traffic sign detection and classification for driver assistant system (DAS). Color detection framework using RGB method is utilized in this study, whereas an artificial neural network (ANN) has been used as classifiers for classification. There are at least 100 types of Malaysian Traffic Signs have been employed in this research. Most of the images are taken at various places throughout the urban and suburban areas involved with scale, illumination and rotational changes as well as occlusion images. The experimental results are shown that the proposed framework achieved at least 80 % successful detection with 21 false positive images. On the other hand, the ANN gives strong rates where at least most of the signs can be classify with more than 85 % success.

Hidden nodes of neural network: Useful application in traffic sign recognition

In this paper we presented a technique to extend the supervised feed forward back propagation neural network that we have previously implemented using out-of-plane traffic sign recognition. In this method we have designed a method to find the optimal number of input nodes together with the hidden nodes to acquire the best system performance. This method not only is able to present the proper combinations between the number of input nodes and hidden layers, but also it can train the network to the optimum stage in the shortest possible time. The result is later plotted and the values of input classes and hidden nodes that give the MSE less that 0.01 is found to be 12 and 55, respectively.

Performance comparison between ANN and PCA techniques for road signs recognition

This study reports about a comparison in recognizing road signs between Neural Network and Principal Component Analysis (PCA). The road sign with circular, triangular, octagonal and diamond shapes have been used in this study. Two recognition systems to determine the classes of the road signs class were implemented which are based on Feed Forward Neural Network and Principal Component Analysis (PCA). The performance of the trained classifier using Scaled Conjugate Gradient (SCG) back propagation function in Neural Network and PCA technique were evaluated on our test datasets. The experiments show that the system using PCA has a higher accuracy as compared to Neural Network with a minimum of 94% classification rate of road signs.

Object classification and recognition using Bag-of-Words (BoW) model

This paper presents the findings on Object Recognition using Bag-of-words model done by the author. The introduction section covers the basic introduction to image processing and object recognition. Details description on the flow of experiment conducted as part of the research will be given in the methodology section. Clear explanation on Bag-of-Words model and Confusion Matrix is included in section III and section IV respectively. Finally, a preliminary result is shown together explanation and a brief conclusion is made at the end of this journal.

Performance analysis of robust road sign identification

This study describes performance analysis of a robust system for road sign identification that incorporated two stages of different algorithms. The proposed algorithms consist of HSV color filtering and PCA techniques respectively in detection and recognition stages. The proposed algorithms are able to detect the three standard types of colored images namely Red, Yellow and Blue. The hypothesis of the study is that road sign images can be used to detect and identify signs that are involved with the existence of occlusions and rotational changes. PCA is known as feature extraction technique that reduces dimensional size. The sign image can be easily recognized and identified by the PCA method as is has been used in many application areas. Based on the experimental result, it shows that the HSV is robust in road sign detection with minimum of 88% and 77% successful rate for non-partial and partial occlusions images. For successful recognition rates using PCA can be achieved in the range of 94–98%. The occurrences of all classes are recognized successfully is between 5% and 10% level of occlusions.

Local Binary Pattern (LBP) with application to variant object detection: A survey and method

This paper study about variants object detection by using local binary pattern. Local binary pattern is one of the famous method in object detection field because of its success used in object detection. The objective of object detection is to differentiate between object and background. However, LBP also has its own weaknesses in object detection. LBP modification that been proposed by a lot of researchers can overcome the weaknesses. In this paper, variants of local binary pattern method and modification has been study and analyze. All those local binary pattern modification has been extract its feature in term of object detection. The modifications are Non-Redundant Local Binary Pattern (NRLBP), Integral Local Binary Pattern (INTLBP), Multi-scale Block Local Binary Pattern (MBLBP) and Discriminative Robust Local Binary Pattern (DRLBP).

Design and Develop an Autonomous UAV Airship for Indoor Surveillance and Monitoring Applications

This project is about to develop an airship based on small size remotely controlled by human. The airship is one of Unmanned Airship Vehicle (UAV) which is can be apply in advertising, VIP security inspection, traffic monitoring and management and so on. The main purpose of this project is to design and develop an autonomous UAV airship for indoor surveillance and monitoring applications. The image will be captured from wireless camera where it mounted at a bottom of gondola. To determine the centroid points of the object are implemented in three phase edge detector, canny operator and threshold. The object will be display on Graphical User Interface (GUI) in 2D coordinated. In this project the systems able to detect only one object at one time.

An analysis on out-of plane face detection among female student and illumination effects using Sift and Surf

This paper discusses the implementation of Scale-Invariant Feature Transform (SIFT) and Speeded-Up Robust Features (SURF) in order to analyse their performance. The problem of image processing which are out-of-plane image and the illumination effect` was analysed by using both techniques. The simulation of both techniques were developed in Matlab 2013a to verify the response performance of the techniques. Research was conducted in order to get the result on the SURF key point detected, the matching key point and time response of the technique performance. The results of the implementation of SURF technique showed that SURF is faster than other comparable techniques.

Analysis of frontal face detection performance by using Artificial Neural Network (ANN) and Speed-Up Robust Features (SURF) technique

The limitation of surveillance camera (CCTV) is related to the low resolution of the camera. In face detection, low resolution will affect the recognition rates and performance of the algorithm in terms of time response and its accuracy. This report leads to an Analysis of Frontal Face Detection by using Artificial Neural Network (ANN) and Speeded-Up Robust Feature (SURF) technique. The implementation of frontal face detection by using two varied techniques of image processing of Neural Network and SURF technique will be explored by using MATLAB software. Both techniques generate contrast of image performance in terms of time response and its accuracy. The expected output is to generate frontal face detection and to compare the performance of the image between the techniques selected. The comparison of performance in terms of time response, SURF is much better than ANN. This can be seen clearly by varying the resolution of the image. SURF keeps the fast record in identify the key feature in the image. In ...