2 research outputs found
Iraq After Saddam Hussein and the Future of Southern Kurdistan, Duhok [Elektronisk resurs] : Hawar, 2008
No full textThis book is an anthology of a number of inquiries regarding Iraq after the fall of the Baath regime in Iraq in 2003, written by the author. There are a range of themes been discussed in the book, the Kurds in Iraq at crossroads, the process of rebuilding the state of Iraq, the US foreign policy and the case of Iraq, regional power and the rebuilding of Iraq, Turkey’s policies vis a vis Iraq and the Kurdish administration in that country, … etc. There are also a number of interviews conducted with the author, republished in this anthology
Plate Number Recognition based on Hybrid Techniques
No full textGlobally and locally, the number of vehicles is on the rise. It is becoming more and more challenging for authorities to track down specific vehicles. Automatic License Plate Recognition becomes an addition to transportation systems automation. Where the extraction of the vehicle license plate is done without human intervention. Identifying the precise place of a vehicle through its license plate number from moving images of the vehicle image is among the crucial activities for vehicle plate discovery systems. Artificial intelligence systems are connecting the gap between the physical world and digital world of automatic license plate detection. The proposed research uses machine learning to recognizing Arabic license plate numbers. An image of the vehicle number plate is captured and the detection is done by image processing, character segmentation which locates Arabic numeric characters on a number plate. The system recognizes the license plate number area and extracts the plate area from the vehicle image. The background color of the number plate identifies the vehicle types: (1) White color for private vehicle; (2) red color for bus and taxi; (3) blue color for governmental vehicle; (4) yellow color for trucks, tractors, and cranes; (5) black color for temporary license; and (6) green color for army. The recognition of Arabic numbers from license plates is achieved by two methods as (1) Google Tesseract OCR based recognition and (2) Machine Learning-based training and testing Arabic number character as K-nearest neighbors (kNN). The system has been tested on 90 images downloaded from the internet and captured from CCTV. Empirical outcomes show that the proposed system finds plate numbers as well as recognizes background color and Arabic number characters successfully. The overall success rates of plate localization and background color detection have been done. The overall success rate of plate localization and background color detection is 97.78%, and Arabic number detection in OCR is 45.56 % as well as in KNN is 92.22%
