Car make and model recognition (CMMR) has become an important part of intelligent transport systems. Information provided by CMMR can be utilized when license plate numbers cannot be identified or fake number plates are used. CMMR can also be used when a certain model of a vehicle is required to be automatically identified by cameras. The majority of existing CMMR methods are designed to be used only in daytime when most of the car features can be easily seen. Few methods have been developed to cope with limited lighting conditions at night where many vehicle features cannot be detected. The aim of this work was to identify car make and model at night by using available rear view features. This paper presents a one-class classifier ensemble designed to identify a particular car model of interest from other models. The combination of salient geographical and shape features of taillights and license plates from the rear view is extracted and used in the recognition process. The majority vote from support vector machine, decision tree, and k-nearest neighbors is applied to verify a target model in the classification process. The experiments on 421 car makes and models captured under limited lighting conditions at night show the classification accuracy rate at about 93 %.

Vehicle detection is an important process of many advance driver assistance system (ADAS) such as forward collision avoidance, Time to collision (TTC) and Intelligence headlight control (IHC). This paper presents a new algorithm to detect a vehicle ahead by using taillight pair. First, the proposed method extracts taillight candidate regions by filtering taillight colour regions and applying morphological operations. Second, pairing each candidates and pair symmetry analysis steps are implemented in order to have taillight positions. The aim of this work is to improve the accuracy of taillight detection at night with many bright spot candidates from streetlamps and other factors from complex scenes. Experiments on still images dataset show that the proposed algorithm can improve the taillight detection accuracy rate and robust under limited light images. General TermsThe general term of this study is image processing and computer vision.

Abstract-This paper presents an automated ontology framework for service robots. The framework is designed to automatically create an ontology and an instance of concept in dynamic environment. Ontology learning from text is applied to build a concept hierarchy using WordNet which provides a rich semantic processing for physical objects. The Automated Ontology is composed of four modules: Concept Creation, Property Creation, Relationship Creation and Instance of Concept Creation. The automated ontology algorithm was implemented in order to create the concept hierarchy in the Robot Ontology. The Semantic Knowledge Acquisition represents knowledge of physical objects in dynamic environments. In simulation experiments, the list of object names and property names was identified. The result shows the concept hierarchy which represents explicit terms and the semantic knowledge of physical objects for performing everyday manipulation tasks.

Early diagnosis of colorectal cancer focuses on detecting polyps in the colon as early as possible so that patients can have the best chances for success- ful treatment. This research presents the optimized parameters for polyp detection using a deep learning technique. Polyp and non-polyp images are trained on the InceptionResnetV2 model by the Faster Region Con- volutional Neural Networks (Faster R-CNN) framework to identify polyps within the colon images. The proposed method revealed more remarkable results than previous works, precision: 92.9 %, recall: 82.3%, F1-Measure: 87.3%, and F2-Measure: 54.6% on public ETIS-LARIB data set. This detection technique can reduce the chances of missing polyps during a pro- longed clinical inspection and can improve the chances of detecting multiple polyps in colon images.

License plate localization is considered to be the significant process in Automatic Number Plate Recognition (ANPR) system, because the accuracy rate of license plate recognition relies on the performance of license plate localization. The majority of license plate localization papers are dedicated to daytime where many appearances can be used to locate license plate. The researches were also reported with high detection accuracy, more than 90%. However, a few studies are presented at night when license plate appearances are not easy to obtain. In this condition, license plate detection is very challenging due to the limitation of available appearances and other light sources may interfered. This paper presents a method to detect license plate position at night by combining color-based, edge-based and image processing techniques. The technique uses a variety of sizes of sub-image to improve local contrast in order to solve problems of low contrast and uneven-light images. The experiments were conducted on images at night in various lighting conditions and the method can detect license plate position with accuracy rate of about 85%.