Jules-Raymond Tapamo

Maritime surveillance

Research highlights? Existing vision-based tracking methods are not suitable for the maritime domain. ? We derive a suitable tracking method by combining and modifying existing methods. ? Our method can track tiny targets. ? Our method is validated on several test sequences and two live field trials. Surveillance in a maritime environment is indispensable in the fight against a wide range of criminal activities, including pirate attacks, unlicensed fishing trailers and human trafficking. Computer vision systems can be a useful aid in the law enforcement process, by for example tracking and identifying moving vessels on the ocean. However, the maritime domain poses many challenges for the design of an effective maritime surveillance system. One such challenge is the tracking of moving vessels in the presence of a moving dynamic background (the ocean). We present techniques that address this particular problem. We use a background subtraction method and employ a real-time approximation of level-set-based curve evolution to demarcate the outline of moving vessels in the ocean. We report promising results on both small and large vessels, based on two field trials.

Integrating Iris and Signature Traits for Personal Authentication Using User-Specific Weighting

Biometric systems based on uni-modal traits are characterized by noisy sensor data, restricted degrees of freedom, non-universality and are susceptible to spoof attacks. Multi-modal biometric systems seek to alleviate some of these drawbacks by providing multiple evidences of the same identity. In this paper, a user-score-based weighting technique for integrating the iris and signature traits is presented. This user-specific weighting technique has proved to be an efficient and effective fusion scheme which increases the authentication accuracy rate of multi-modal biometric systems. The weights are used to indicate the importance of matching scores output by each biometrics trait. The experimental results show that our biometric system based on the integration of iris and signature traits achieve a false rejection rate (FRR) of 0.08% and a false acceptance rate (FAR) of 0.01%.

Adaptive Thresholding Technique for Retinal Vessel Segmentation Based on GLCM-Energy Information

Although retinal vessel segmentation has been extensively researched, a robust and time efficient segmentation method is highly needed. This paper presents a local adaptive thresholding technique based on gray level cooccurrence matrix- (GLCM-) energy information for retinal vessel segmentation. Different thresholds were computed using GLCM-energy information. An experimental evaluation on DRIVE database using the grayscale intensity and Green Channel of the retinal image demonstrates the high performance of the proposed local adaptive thresholding technique. The maximum average accuracy rates of 0.9511 and 0.9510 with maximum average sensitivity rates of 0.7650 and 0.7641 were achieved on DRIVE and STARE databases, respectively. When compared to the widely previously used techniques on the databases, the proposed adaptive thresholding technique is time efficient with a higher average sensitivity and average accuracy rates in the same range of very good specificity.

Detection and tracking of moving objects in a maritime environment using level set with shape priors

Over the years, maritime surveillance has become increasingly important due to the recurrence of piracy. While surveillance has traditionally been a manual task using crew members in lookout positions on parts of the ship, much work is being done to automate this task using digital cameras coupled with a computer that uses image processing techniques that intelligently track object in the maritime environment. One such technique is level set segmentation which evolves a contour to objects of interest in a given image. This method works well but gives incorrect segmentation results when a target object is corrupted in the image. This paper explores the possibility of factoring in prior knowledge of a ship’s shape into level set segmentation to improve results, a concept that is unaddressed in maritime surveillance problem. It is shown that the developed video tracking system outperforms level set-based systems that do not use prior shape knowledge, working well even where these systems fail.

Improving Iris-Based Personal Identification Using Maximum Rectangular Region Detection

Iris recognition is proving to be one of the most reliable biometric traits for personal identification. In fact, iris patterns have stable, invariant and distinctive features for personal identification. In this paper, we propose a new algorithm that detects the largest non-occluded rectangular part of the iris as region of interest (ROI). Thereafter, a cumulative-sum-based grey change analysis algorithm is applied to the ROI to extract features for recognition. This method could possibly be utilized for partial iris recognition since it relaxes the requirement of using the whole part of the iris to produce an iris template. Preliminary experimental results carried on a CASIA iris database, show that the approach is promisingly effective and efficient.

Comparative Study of Retinal Vessel Segmentation Based on Global Thresholding Techniques

Due to noise from uneven contrast and illumination during acquisition process of retinal fundus images, the use of efficient preprocessing techniques is highly desirable to produce good retinal vessel segmentation results. This paper develops and compares the performance of different vessel segmentation techniques based on global thresholding using phase congruency and contrast limited adaptive histogram equalization (CLAHE) for the preprocessing of the retinal images. The results obtained show that the combination of preprocessing technique, global thresholding, and postprocessing techniques must be carefully chosen to achieve a good segmentation performance.

An active contour approach to insulator segmentation

Faulty insulator causes the change of voltage and electric current, which generates great damage in power grid. It is therefore important to monitor and inspect the insulator to assess damage that could be caused by aging or any accident on power line system. Computer vision has been identified and is being investigated as a tool to solve this problem more safely, accurately and speedily. Identifying the region where the insulator is located is crucial in the process of assessing its status. The insulator can then be segmented and fed to defect detector. In this paper we proposed an insulator segmentation framework from plain and complex background using active contour model and extracting insulator region of interest. Experiment shows that active contour model successfully segment insulators from a plain and complex background by evaluating the result with a manually created ground-truth. This method is more efficient, flexible, and outperforms the classical methods such as thresholding and gradients based approach.

A New Adaptive Thresholding Technique for Retinal Vessel Segmentation Based on Local Homogeneity Information

Segmentation of vessels in retinal images has become challenging due to the presence of non-homogeneous illumination across retinal images. This paper develops a novel adaptive thresholding technique based on local homogeneity information for Retinal vessel segmentation. Different types of local homogeneity information were investigated. An experimental evaluation on DRIVE database demonstrates the high performance of all types of homogeneity considered. An average accuracy of 0.9469 and average sensitivity of 0.7477 were achieved. While compared with widely previously used techniques on DRIVE database, the proposed adaptive thresholding technique is superior, with a higher average sensitivity and average accuracy rates in the same range of very good specificity.

Signature Verification Based on Handwritten Text Recognition

Signatures continue to be an important biometric trait because it remains widely used primarily for authenticating the identity of human beings. This paper presents an efficient text-based directional signature recognition algorithm which verifies signatures, even when they are composed of special unconstrained cursive characters which are superimposed and embellished. This algorithm extends the character-based signature verification technique. The experiments carried out on the GPDS signature database and an additional database created from signatures captured using the ePadInk tablet, show that the approach is effective and efficient, with a positive verification rate of 94.95%.

Handwritten signature verification using weighted fractional distance classification

Signatures are one of the behavioural biometric traits, which are widely used as a means of personal verification. Therefore, they require efficient and accurate methods of authenticating users. The use of a single distance-based classification technique normally results in a lower accuracy compared to supervised learning techniques. This paper investigates the use of a combination of multiple distance-based classification techniques, namely individually optimized re-sampling, weighted Euclidean distance, fractional distance and weighted fractional distance. Results are compared to a similar system that uses support vector machines. It is shown that competitive levels of accuracy can be obtained using distance-based classification. The best accuracy obtained is 89.2%.