Hashem Al-Nabhi | Northwestern Polytechnical University (original) (raw)

Uploads

Papers by Hashem Al-Nabhi

International Journal of Computing and Digital Systems

Defective Characters exist frequently and broadly in images such as license plates, electricity, ... more Defective Characters exist frequently and broadly in images such as license plates, electricity, water meters, street boards, etc. Thus, building robust recognition systems or enhancing the accuracy and robustness of the existing recognition systems to recognize such characters on images is a challenging research topic in image processing and computer vision. This paper Investigates and adopts ReId dataset for all the experimental work and introduces two deep learning models (CNN5-BLSTM and CNN7-GRU) based on convolutional recurrent neural networks (CRNN) to address the problem of defective characters sequence recognition. The two proposed deep learning models are segmentation-free, lightweight, End-To-End trainable, and slightly different from each other. The models are evaluated on testing data of ReId dataset, and the achieved accuracies are 95% of characters' sequence accuracy and 98% of character-level accuracy. Moreover, their performance on ReId dataset outperforms other models' performance in the literature. .

Defective Characters exist frequently and broadly in images such as license plates, electricity, ... more Defective Characters exist frequently and broadly in images such as license plates, electricity, water meters, street boards, etc. Thus, building robust recognition systems or enhancing the accuracy and robustness of the existing recognition systems to recognize such characters on images is a challenging research topic in image processing and computer vision. This paper Investigates and adopts ReId dataset for all the experimental work and introduces two deep learning models (CNN5-BLSTM and CNN7-GRU) based on convolutional recurrent neural networks (CRNN) to address the problem of defective characters sequence recognition. The two proposed deep learning models are segmentation-free, lightweight, End-To-End trainable, and slightly different from each other. The models are evaluated on testing data of ReId dataset, and the achieved accuracies are 95% of characters' sequence accuracy and 98% of character-level accuracy. Moreover, their performance on ReId dataset outperforms other models' performance in the literature. .

This paper presents an enhanced methodology in implementing and designing a security system for d... more This paper presents an enhanced methodology in implementing and designing a security system for door locking purpose based on fingerprint, GSM technology, monitoring camera, alarm system and password system. This security system will provide enough security by limiting unauthorized people access and taking a record of those who pass through it. Sometimes unauthorized people or burglars try to break the door for evil intentions at a time when no one is available at a targeted place, so this paper introduces some security solutions for that problem and they are the main contribution of our paper. We introduce an alarm system to alert the people at the surroundings, GSM module that's used to send an SMS message to the registered user's (responsible person) and a web camera that's used to take a video for a person who tries to break the lock, password keypad that's used after fingerprint sensing to provide extra security. Definitely the registered users are the only persons who can access the lock, and the door closes after five seconds from the opening time. The method used to implement this experiment involves the use of a fingerprint scanner R305 that's interfaced with Arduino microcontroller-ATMEGA328P to control the locking and unlocking process of a door. During all the opening and closing processes, the16x2 Liquid Crystal Display (LCD) displays some commands which can be used to instruct the users like, place your finger on the sensor, the door is opened, the door is closed, the message is sent, please enter the password etc. If an unregistered user tries to access the door using their fingerprints, automatically his/her access is denied. The proposed door lock security system is can be used at homes, offices, banks, hospitals, and in other governmental and private sectors. Our proposed system was tested in real-time and has shown competitive results compared to other projects using RFI and password.

International Journal of Computing and Digital Systems

Defective Characters exist frequently and broadly in images such as license plates, electricity, ... more Defective Characters exist frequently and broadly in images such as license plates, electricity, water meters, street boards, etc. Thus, building robust recognition systems or enhancing the accuracy and robustness of the existing recognition systems to recognize such characters on images is a challenging research topic in image processing and computer vision. This paper Investigates and adopts ReId dataset for all the experimental work and introduces two deep learning models (CNN5-BLSTM and CNN7-GRU) based on convolutional recurrent neural networks (CRNN) to address the problem of defective characters sequence recognition. The two proposed deep learning models are segmentation-free, lightweight, End-To-End trainable, and slightly different from each other. The models are evaluated on testing data of ReId dataset, and the achieved accuracies are 95% of characters' sequence accuracy and 98% of character-level accuracy. Moreover, their performance on ReId dataset outperforms other models' performance in the literature. .

Defective Characters exist frequently and broadly in images such as license plates, electricity, ... more Defective Characters exist frequently and broadly in images such as license plates, electricity, water meters, street boards, etc. Thus, building robust recognition systems or enhancing the accuracy and robustness of the existing recognition systems to recognize such characters on images is a challenging research topic in image processing and computer vision. This paper Investigates and adopts ReId dataset for all the experimental work and introduces two deep learning models (CNN5-BLSTM and CNN7-GRU) based on convolutional recurrent neural networks (CRNN) to address the problem of defective characters sequence recognition. The two proposed deep learning models are segmentation-free, lightweight, End-To-End trainable, and slightly different from each other. The models are evaluated on testing data of ReId dataset, and the achieved accuracies are 95% of characters' sequence accuracy and 98% of character-level accuracy. Moreover, their performance on ReId dataset outperforms other models' performance in the literature. .

This paper presents an enhanced methodology in implementing and designing a security system for d... more This paper presents an enhanced methodology in implementing and designing a security system for door locking purpose based on fingerprint, GSM technology, monitoring camera, alarm system and password system. This security system will provide enough security by limiting unauthorized people access and taking a record of those who pass through it. Sometimes unauthorized people or burglars try to break the door for evil intentions at a time when no one is available at a targeted place, so this paper introduces some security solutions for that problem and they are the main contribution of our paper. We introduce an alarm system to alert the people at the surroundings, GSM module that's used to send an SMS message to the registered user's (responsible person) and a web camera that's used to take a video for a person who tries to break the lock, password keypad that's used after fingerprint sensing to provide extra security. Definitely the registered users are the only persons who can access the lock, and the door closes after five seconds from the opening time. The method used to implement this experiment involves the use of a fingerprint scanner R305 that's interfaced with Arduino microcontroller-ATMEGA328P to control the locking and unlocking process of a door. During all the opening and closing processes, the16x2 Liquid Crystal Display (LCD) displays some commands which can be used to instruct the users like, place your finger on the sensor, the door is opened, the door is closed, the message is sent, please enter the password etc. If an unregistered user tries to access the door using their fingerprints, automatically his/her access is denied. The proposed door lock security system is can be used at homes, offices, banks, hospitals, and in other governmental and private sectors. Our proposed system was tested in real-time and has shown competitive results compared to other projects using RFI and password.