A Meta-Analysis on Obstacle Detection for Visually Impaired People

Volume 6 Issue 1 March - May 2019

Survey Paper

A Meta-Analysis on Obstacle Detection for Visually Impaired People

N. Veeranjaneyulu*, K. K. Baseer**, V. S. Asha***, T. Madhu Prakash****

*_****Department of Information Technology, Sree Vidyanikethan Engineering College, Tirupati, India.

Veeranjaneyulu, N., Baseer, K., K., Asha, V., S., Madhu Prakash, T. (2019). A Meta-Analysis on Obstacle Detection for Visually Impaired People.i-manager’s Journal on Pattern Recognition, 6(1), 40-62. https://doi.org/10.26634/jpr.6.1.15523

Abstract

In general humans have five senses, among all vision is the most important and best gift given to the humans by GOD, but it is limited to some of the people due to their Visual Impairment issues. If vision is the problem then GOD will give the capabilities in other senses. The proportion of visually impaired and blind people in the overall world has become a very large. In a survey report given by WHO (World Health Organization) in 2010, they estimated nearly 285.389 million people are suffering with visual impairment problems across the globe. Many equipment's (Ex: Cane, Assistive shoe, Spectacles) are developed by different authors for detection of obstacles by visual impaired people over the time. All these equipment's are developed by using different techniques like IoT enabled smart cane, GPS/GSM based smart cane, Wearable devices like Assistive shoe's and blind vision spectacles which detects the obstacles, Smart Phone based navigation technology , Image processing techniques based smart cane which uses the camera for capturing the images, ETA's (Electronic Travel Aid's), normal Ultrasonic sensor based smart canes, Sensors(Ultrasonic, LDR's, Soil moisture and water detection) used smart cane and the most advanced smart canes which uses the Algorithms of Machine Learning and Deep Learning ANN, CNN, RNN. In this paper, we present a clear survey of the navigation systems of blind/Visual impaired people that are proposed by different authors highlighting various technologies used, designs implemented, working challenges faced and requirements of blind people for their autonomous navigation either in indoor or outdoor environment. Also we aims at presenting several existing literatures which are based on object detection by blind people. Due to the advancement in techniques and technology, study, analysis and evaluation of all these proposals by different authors will play a vital role. Hence this survey will concentrate on analyzing the process involved in detection of obstacles with different techniques.

Overview of Motion Estimation Algorithms for Video Coding

Volume 6 Issue 1 March - May 2019

Survey Paper

Overview of Motion Estimation Algorithms for Video Coding

Kiran Kumar Vemula*, Neeraja S.**

*Department of Electronics and Communication Engineering, Malla Reddy College of Engineering and Technology, Hyderabad, Telangana, India.

** Department of Electronics and Communication Engineering, GITAM Institute of Technology, Visakhapatnam, Andhra Pradesh, India.

Vemula, K., K., Neeraja. S. (2019). Overview of Motion Estimation Algorithms for Video Coding.i-manager’s Journal on Pattern Recognition, 6(1), 35-39. https://doi.org/10.26634/jpr.6.1.16511

Abstract

Motion estimation process is an important module in digital video coding applications as it demands more computations when compared to other modules of digital video coding. In order to overcome this difficulty, many motion estimation algorithms were proposed. This paper presents an analysis of some famous algorithms in motion estimation process for digital video coding. In this review, the search procedures, computational complexity and quality of these algorithms are discussed.

Estimation of volume of a solid object From Three Dimensional Point Clouds Generated By Convolutional Neural Networks based Semantic Segmentation

Volume 6 Issue 1 March - May 2019

Research Paper

Estimation of volume of a solid object From Three Dimensional Point Clouds Generated By Convolutional Neural Networks based Semantic Segmentation

Radhamadhab Dalai*

* Department of Computer Science & Engineering,Birla Institute of Technology, Ranchi, India.

Dalai, R. (2019). Estimation of volume of a solid object From Three Dimensional Point Clouds Generated By Convolutional Neural Networks based Semantic Segmentation.i-manager’s Journal on Pattern Recognition, 6(1), 27-34. https://doi.org/10.26634/jpr.6.1.16458

Abstract

Generating three dimensional point cloud for an object in image has found many applications in used in many computer vision systems. In this work a convolutional neural network based semantic segmentation has been used to find region of interest in an image. The region of interest has been represented as point clouds in three dimensional space. Then using image processing technique area based filter operations have been applied to find the total surface area. Finally adding all these small volumes total volume has been calculated. A large number of algorithms have been adapted reconstruction methods have been experimented and tested only for uniform backgrounds, which is disadvantageous for the applications on real images which consists of complex nonuniform regions. In this work semantic segmentation has been used to partition the regions into similar instance based regions. We have used UNET model for the region based segmentation. Then using encoderdecoder scheme the 3D point cloud has been generated after merging pixel clouds. This paper proposes an end-to-end efficient generation network, which is composed of an encoder, a 3D image model, and a decoder. First, a single-view image of object and a nearest-shape retrieval has been formed from UNET are fed into the network; then, the two encoders are merged adaptively according to their homo-graphic or similarity in nature. Then decoder generates fine-grained point clouds from the pixel clouds generated from multiple view images. Each point in the cloud represents a weight according the intensity and color information from which the density and volume of object has been calculated. The experiments on uniform background images show that our method attains accuracy 12 to 15 %margin compared with volumetric and point set generation methods particularly toward large solid objects, and it works multiple view angles as well.

Robot Control using Hand Gesture

Volume 6 Issue 1 March - May 2019

Research Paper

Robot Control using Hand Gesture

U. B. Mahadewaswamy*, Anusha H. N**

*-** Department of Electronics and Communication, JSSS & TU, Mysuru, India.

Mahadevaswamy , U., B., Anusha, H., N. (2019). Overview Robot Control Using Hand Gesture.i-manager’s Journal on Pattern Recognition, 6(1), 11-26. https://doi.org/10.26634/jpr.6.1.15963

Abstract

Hand controllers and electromechanical devices have been used by humans to control robots or machines but there were some constraints in several factors of interaction. Pattern recognition and Gesture recognition are the growing fields of analysis. Hand gesture recognition is very significant for human-computer interaction (HCI). In this work, we present a completely unique real-time methodology for robot control using hand gesture recognition. It is necessary for the user to communicate and control a device in the natural efficient way in human-robot interaction based. The implementation is done using Kinect sensor and Matlab environment. The robot arm is controlled by Firebird V robot. We have implemented a prototype using gesture as a tool for communication with ma-chine command signals are generated using gesture control algorithm. These generated signals are then given to the robot to perform a set of task. This Kinect sensor recognizes the hand gestures and then assigns functions to be performed by the robot for each hand gesture.

Artificial Neural Network-Based Pelvic Inflammatory Disease Diagnosis System

Volume 6 Issue 1 March - May 2019

Research Paper

Artificial Neural Network-Based Pelvic Inflammatory Disease Diagnosis System

Yahaya Mohammed Sani*, Dere Boluwatife Adesola**, Hussaini Abubakar Zubairu***, Ilyasu Anda****

*-*** Department of Information and Media Technology, Federal University of Technology, Minna, Nigeria.

**** Department of Library and Information Technology, Federal University of Technology, Minna, Nigeria.

Sani , Y., M., Adesola, D., B., Zubairu, H., A., & Anda, I. (2019). Artificial Neural Network-Based Pelvic Inflammatory Disease Diagnosis System. i-manager’s Journal on Pattern Recognition, 6(1), 1-10. https://doi.org/10.26634/jpr.6.1.16510

Abstract

Pelvic Inflammatory Disease (PID) is a reproductive health infective disease of feminine genital tract and is commonly affecting the young women and adult female. Clinical manifestation of PID differs among patients and decision of medical experts are based on clinician experience instead of hidden data in the knowledge database. The diagnosis of PID based on heuristic lead to errors, where ectopic pregnancy could be mistaken for PID. This paper presents Artificial Neural Network based model to diagnose pelvic inflammatory diseases based on a set of clinical data. The ANN model was trained with 259 clinical data as input to the neural network. The system can predict the presence or absence of PID based on the available symptoms. An accuracy of 96.1% was recorded based on the confusion matrix. The obtained result is promising, an indication that the system can be effective in diagnosis of PID cases.

Separation, Classification and Expert Mapping of Old Grantha Documents Symbols

Volume 5 Issue 4 December - February 2019

Research Paper

Separation, Classification and Expert Mapping of Old Grantha Documents Symbols

Lalit Prakash Saxena*

* Research Scientist, Applied Research Section, Combo Consultancy, Obra UP India.

Saxena, L. P. (2019). Separation, Classification and Expert Mapping of Old Grantha Documents Symbols. i-manager’s Journal on Pattern Recognition, 5(4), 51-67. https://doi.org/10.26634/jpr.5.4.16108

Abstract

This paper attempts to decipher old documents using symbol to script mapping scheme. Symbols are confined to documents either as isolated notations or handwritten texts with a number of not able features. This paper describes a method to separate and classify handwritten non-cursive symbols in Grantha script. This work uses statistical correlation coefficient method for separation and classification, without the recognition of the symbols. The Grantha script symbols mapping model comprises of selection, separation, preprocessing, classification, and finally mapping. The proposed model employs bounding box algorithm for locating the symbols. The algorithm selects the symbols and excludes the non-symbol components to an extent possible. For experiments, 135 Grantha script document images of varying deteriorating complexities were used. The resulting symbol classification rate (i.e., the proportion of symbols automatically classified) was obtained near to 80%, aiding in mapping to a predetermined mapping scheme.

A Medical Expert System for Predicting the Prevalence of Autoimmune Diabetes Mellitus in Thyroid Patients

Volume 5 Issue 4 December - February 2019

Research Paper

A Medical Expert System for Predicting the Prevalence of Autoimmune Diabetes Mellitus in Thyroid Patients

Surekha Samsani *, G.Jaya Suma**

* Department of Computer Science and Engineering, UCEK(A), Jawaharlal Nehru Technological University Kakinada, Andhra Pradesh, India.

** Department of Information Technology, UCEV(A), Jawaharlal Nehru Technological University Kakinada, Andra Pradesh, India.

Samsani,S.,&Suma,G.J. (2019). A Medical Expert System for Predicting the Prevalence of Autoimmune Diabetes Mellitus in Thyroid Patients. i-manager’s Journal on Pattern Recognition, 5(4), 44-50. https://doi.org/10.26634/jpr.5.4.15947

Abstract

Diabetes Mellitus (DM) and Thyroid are the major coexistent autoimmune disorders affecting people globally. Due to prolonged chronic mental stress in the modern lifestyle, Thyroid disorder is affecting all age groups and people with Thyroid disorder have an increased risk of developing DM complications. Because, abnormal Thyroid dysfunction can have dreadful effects on blood glucose control and can affect the course of DM. This paper proposes a Medical Expert system to assist clinicians in predicting the prevalence of developing autoimmune DM more precisely in patients suffering from Thyroid and further helps to investigate better in the line of improving public health. In this work, Fuzzy logic based inference system and unsupervised machine learning algorithms are used to discover associations and dependencies between Thyroid and DM. The inferred knowledge base is used to design Fuzzy based Expert system. To develop a more realistic expert system, blood sample reports of people affected by DM and Thyroid disorder have been collected from various Endocrine centres in Andhra Pradesh, India. Specificity, Sensitivity, Predictive Values, and Likelihood Ratios of the proposed system are promising in support of system functionality

Profiling Inappropriate Users’ Tweets Using Deep Long Short-Term Memory (LSTM) Neural Network

Volume 5 Issue 4 December - February 2019

Research Paper

Profiling Inappropriate Users’ Tweets Using Deep Long Short-Term Memory (LSTM) Neural Network

Abubakar Umar*, Sulaimon A. Bashir**, Laud Charles Ochei***, Ibrahim A. Adeyanju****

*-** Department of Computer Science, Federal University of Technology Minna Nigeria.

*** Robert Gordon University, Aberdeen, UK.

**** Department of Computer Engineering, Federal University, Oye-Ekiti, Nigeria.

Umar, A., Bashir, S. A., Ochei, L.C., & Adeyanju, I. A. (2019). Profiling Inappropriate Users' Tweets Using Deep Long Short-Term Memory (LSTM) Neural Network. i-manager’s Journal on Pattern Recognition, 5(4), 27-43. https://doi.org/10.26634/jpr.5.4.15864

Abstract

In recent times, big Internet companies have come under increased pressure from governments and NGOs to remove inappropriate materials from social media platforms (e.g., Twitter, Facebook, YouTube). A typical example of this problem is the posting of hateful, abusive, and violent tweets on Twitter which has been blamed for inciting hatred, violence and causing societal disturbances. Manual identification of such tweets and the people who post these tweets is very difficult because of the large number of active users and the frequency with which such tweets are posted. Existing approaches for identifying inappropriate tweets have focused on the detection of such tweets without identifying the users who post them. This paper proposes an approach that can automatically identify different types of inappropriate tweets together with the users who post them. The proposed approach is based on a user profiling algorithm that uses a deep Long Short-Term Memory (LSTM) based neural network trained to detect abusive language. With the support of word embedding features learned from the training set, the algorithm is able to classify the tweets of users into different abusive language categories. Thereafter, the user profiling algorithm uses the classes assigned to the tweets of each user to profile each user into different abusive language category. Experiments on the test set show that the deep LSTM-based abusive language detection model reached an accuracy of 89.14% on detecting whether a tweet is bigotry, offensive, racist, extremism-related and neutral. Also, the user profiling algorithm obtained an accuracy of 83.33% in predicting whether a user is a bigot, racist, extremist, uses offensive language and neutral.

On the Development of a Novel Smell Agent Optimization (SAO) for Optimization Problems

Volume 5 Issue 4 December - February 2019

Research Paper

On the Development of a Novel Smell Agent Optimization (SAO) for Optimization Problems

A. T. Salawudeen*, M. B. Mu'azu **, Y. A. Sha'aban***, E. A. Adedokun****

*-**,****Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria.

*** Department of Electrical Engineering, University of Hafer Al-Batin, Saudi Arabia.

Salawudeen, A. T., Mu'azu, M. B., Sha'aban, Y. A., & Adedokun, E. A. (2019).On the Development of a Novel Smell Agent Optimization (SAO) For Optimization Problems. i-manager’s Journal on Pattern Recognition, 5(4), 13-26. https://doi.org/10.26634/jpr.5.4.15677

Abstract

This paper presents the development of a new optimization algorithm called the Smell Agent Optimization (SAO). The algorithm uses the phenomenon of smell and the intuitive trailing behavior of an agent to identify a smell source. The developed algorithm has two basic modes used in the optimization process, which are the sniffing mode and trailing mode. In the sniffing mode, the evaporation of smell molecules from a source is modeled and in the trailing mode, the movement of an agent towards the smell molecules is modeled. The performance of SOA was evaluated using 10 benchmark functions and results was compared with PSO, ABC, and GA. Simulation results showed the efficiency of the developed SAO in solving unimodal and multimodal functions.

Human Emotion Recognition from Facial Expressions

Volume 5 Issue 4 December - February 2019

Research Paper

Human Emotion Recognition from Facial Expressions

Ortil Msugh *, Twaki Koko Grace**

*-**Department of Computer Science. FCT College of Education, Abuja, Nigeria.

Msugh, O., & Grace,T. K. (2019). Human Emotion Recognition from Facial Expressions. i-manager’s Journal on Pattern Recognition, 5(4), 1-12. https://doi.org/10.26634/jpr.5.4.15539

Abstract

Emotion recognition remains a potential research area as efforts to make machines to mimic humans in most areas of human life is yet actualized. This paper presents an emotion recognition system, to enhance recognition accuracy for better user experience. Principal Component Analysis (PCA) was implemented via Singular Value Decomposition (SVD) and used for feature extraction process. In classification process, Discrete Hidden Markov Model (HMM) was utilized in a principled manner. Two-dimensional spatial face features were realized by varying quantization levels. The quantization level with the efficient feature description, judged by the highest recognition accuracy was chosen to train the system. The recognition accuracy of the system was studied on two publicly available datasets, namely, JAFFE and Cohn Kanade (CK) datasets. The system showed better performances compared with other state of the art systems.

Analysis on Text Detection and Extraction from Complex Background Images

Volume 5 Issue 3 September - November 2018

Opinion Paper

Analysis on Text Detection and Extraction from Complex Background Images

D. Kavyashree*, T. M. Rajesh**

*-** Assistant Professor, Department of Computer Science and Engineering, SoE, Dayananda Sagar University, Bengaluru, Karnataka, India.

Kavyashree, D., and Rajesh, T. M (2018). Analysis on Text Detection and Extraction from Complex Background Images. i-manager’s Journal on Pattern Recognition, 5(3), 37-43. https://doi.org/10.26634/jpr.5.3.15260

Abstract

Text detection and extraction from the complex images plays a major role in detecting vigorous and valued information. As the rapid growth of obtainable multimedia information and rising prerequisite for data, documentation, indexing and reclamation, many scholars, researchers and scientists have worked a lot on text detection and extraction from the images. The main aim of our work is to give a comparison analysis on the various techniques and methods that were used and applied to detect and extract the text from complex background images. This comparison analysis will help to pick the proper and suitable technique or the method for future purpose. We can find many applications of a text identification and verification such as picture indexing based on text, Image searching the Google based on Keyword, old and required document examination, Extraction of number from number plates of vehicles involved in crime etc. Detecting and extracting the text from images or video is demanding due to unconventionality of textured background, varying font size, different style, resolution, blurring, position, viewing angle and so on. Enormous techniques have already been developed for detecting and extracting the text from the complex background image. All these methods are based on substantial situations. So the purpose of our work is to provide the analysis on the accuracy of widely used algorithms by scholars and researchers in detecting and extracting the text from complex images. In this paper the results of various methods for extracting the text from the images have been analyzed vigorously and this comparison analysis work helps the researches to ease out the time complexity they find in searching for the different combinational works.

Agricultural E-Extension Services: A Hybrid of Multilingual Translation Text-To-Speech-A Framework

Volume 5 Issue 3 September - November 2018

Research Paper

Agricultural E-Extension Services: A Hybrid of Multilingual Translation Text-To-Speech-A Framework

Yahaya Mohammed Sani*, Stella Oluyemi Etuk**, Ilyasu Anda***, Mamman Adamu****

*Assistant Lecturer, Department of Information and Media Technology, Federal University of Technology, Minna, Nigeria.

**Lecturer, Department of Information and Media Technology, School of Information and Communication Technology, Federal University of Technology, Minna, Nigeria.

***Lecturer, Department of Library and Information Technology, Federal University of Technology, Minna, Nigeria.

****Research Scholar, Department of Computer Science, Federal University of Technology, Minna, Nigeria.

Sani, Y. M., Etuk, S. O., Anda, I., and Adamu, M (2018). Agricultural E-Extension Services: A Hybrid of Multilingual Translation Text-To-Speech-A Framework. i-manager’s Journal on Pattern Recognition, 5(3), 29-36. https://doi.org/10.26634/jpr.5.3.15679

Abstract

This paper presents a framework for a text-to-speech translation on Android Devices based on Natural Language Processing (NLP) and text-to-speech synthesizer (TTS) to deliver real-time agricultural update to farmers by agricultural extension service workers (AEW) as speech is the most used and natural way for people to communicate with one another. In order to increase the naturalness of oral communications between Agricultural Extension Service workers and farmers, speech aspects must be involved. This is because most local farmers have good understanding of their local language and have strong preferences for it over any other language. Since, majority of farmers are in rural areas, they have little or no understanding of English language, agriculture research output communicated in English language, may be of little or no use to them, if they are delivered in a foreign language. Text-to-Speech Enabled Hybrid Multilingual Translation framework adopts a serial integration of Natural Language Processing (NLP) on one hand and text-to-speech synthesizer (TTS) interpretation technique using android google translate API text-to-speech synthesizer and recognizer to translate English, Hausa, Yoruba, Ibo and Arabic texts in to speech(es) respectively in accordance with farmers registered dialect on the other hand.

Effects of Data Normalization on Water Quality Model in a Recirculatory Aquaculture System Using Artificial Neural Network

Volume 5 Issue 3 September - November 2018

Research Paper

Effects of Data Normalization on Water Quality Model in a Recirculatory Aquaculture System Using Artificial Neural Network

Taliha A. Folorunso*, Raisa Begum Gul**, Jonathan G. Kolo ***, Suleiman O. E. Sadiku****, Abdullahi M. Orire *****

*Academic Staff, Department of Mechatronics Engineering, Federal University of Technology, Minna, Nigeria.

**Professor, Department of Mechatronics Engineering, Federal University of Technology, Minna, Nigeria.

***Associate Professor, Department of Electrical and Electronics Engineering, Federal University of Technology, Minna, Nigeria.

****Professor, Department of Water Resources, Aquaculture, and Fisheries Technology, Federal University of Technology, Minna, Nigeria.

*****Associate Professor, Department of Water Resources, Aquaculture, and Fisheries Technology, Federal University of Technology, Minna, Nigeria.

Folorunso, T. A., Aibinu, A. M., Kolo, J. G.,Sadiku, S. O. E., and Orire, A. M (2018). Effects of Data Normalization on Water Quality Model in a Recirculatory Aquaculture System Using Artificial Neural Network. i-manager’s Journal on Pattern Recognition, 5(3), 21-28. https://doi.org/10.26634/jpr.5.3.15678

Abstract

Water Quality remains one of the most important factor that influences the aquaculture system as it effects can make or mar the state of organisms as well as the environment. Furthermore, the use of Artificial intelligence especially the Artificial Neural Network (ANN) has greatly improved the forecasting capability of water quality due to better solutions produced as compared to other approaches. The performance of these AI techniques lies in the quality of dataset used for its implementation, which is in turn a function of the preprocessing (Normalization) techniques performed on them. In this paper, the effect of different normalization techniques namely; the Min-Max, Decimal Point, Unitary and the Z-Score were investigated on the prediction of the water quality of the Tank Cultured Re-circulatory Aquaculture System at the WAFT Laboratory, using the ANN. The Water Quality Index was based on the prediction of the Dissolved Oxygen (DO) as a function of the Temperature, Alkalinity, PH and conductivity. The performance of the techniques on the ANN was evaluated using the Mean Square Error (MSE), Nash-Sutcliffe Efficiency coefficient (NSE). The comparison of the evaluation of the various techniques depicts that all the approaches are applicable in the prediction of the DO. The Decimal point technique has the least MSE as compared to others, while the Min-Max technique has better performance with respect to the NSE.

Effect of Feature Ranking on the Detection of Credit Card Fraud: Comparative Evaluation of Four Techniques

Volume 5 Issue 3 September - November 2018

Research Paper

Effect of Feature Ranking on the Detection of Credit Card Fraud: Comparative Evaluation of Four Techniques

John Oloruntoba Awoyemi*, Adebayo Adetunmbi **, Samuel Oluwadare***

* Part-time Lecturer, Department of Computer Science, Federal Polytechnic Ado-Ekiti, Ekiti State, Nigeria.

** Professor, Department of Computer Science, Federal University of Technology, Akure, Nigeria.

*** Senior Lecturer and Head, Department of Computer Science, School of Computing, FUTA, Nigeria.

Awoyemi, J., Adetunmbi, A., and Oluwadare, S (2018). Effect of Feature Ranking on the Detection of Credit Card Fraud: Comparative Evaluation of Four Techniques. i-manager’s Journal on Pattern Recognition, 5(3), 10-20. https://doi.org/10.26634/jpr.5.3.15676

Abstract

Credit card fraud detection is an important aspect of financial institutions that provide various online payment services to its customers. One of criteria which affect performance of credit card fraud detection models is the selection of variables. This paper studies the effects of feature engineering on two sets of feature ranked imbalanced credit card fraud datasets for four classifier techniques. This paper employs the credit card fraud datasets (Taiwan and European bank) obtained from UCI and ULB repositories containing 30,000 and 284,807 transactions respectively. Feature ranking on the sets of datasets is carried out using correlation analysis technique. Algorithms of four classifiers are produced and used on feature and raw ranked data. The algorithms of the classifiers are run in MATLAB. The performance metrics applied in assessing the effects of the four classifiers on the feature and raw ranked datasets are specificity, precision, Matthews correlation coefficient, sensitivity, accuracy, and balanced classification rate. Results from the comparative analysis show that decision tree variants classifiers outperform naïve bayes, support vector and neural network radial basis function techniques respectively. The feature ranked and raw datasets of the European credit card fraud data recorded highest performance metrics for decision trees. The paper investigates the effect of feature ranking of two imbalanced credit card fraud data on four machine learning techniques using filter approach.

Design of a Framework for Computer-Based Examination Invigilation Using Fingerprint and Iris Technologies

Volume 5 Issue 3 September - November 2018

Research Paper

Design of a Framework for Computer-Based Examination Invigilation Using Fingerprint and Iris Technologies

Gabriel Babatunde Iwasokun*, Omomule Taiwo Gabriel **, Rapheal Olufemi Akinyede***

*,*** Senior Lecturer, Department of Computer Science, Federal University of Technology, Akure, Nigeria.

** Assistant Lecturer, Department of Computer Science, Adekunle Ajasin University, Akungba Akoko, Ondo State, Nigeria.

Iwasokun, G. B., Omomule, T. G., and Akinyede, O. R (2018). Design of a Framework for Computer-Based Examination Invigilation Using Fingerprint and Iris Technologies. i-manager’s Journal on Pattern Recognition, 5(3), 1-9. https://doi.org/10.26634/jpr.5.3.15675

Abstract

Computer-Based Examination (CBE) is a new paradigm in the assessment and measurement of knowledge and capabilities. It relies on computer and its associated technologies to provide solution to some of the problems inherent to human-based approaches to conduct of examinations and invigilation. Such problems include connivance, impersonation, external sourcing and peeking. This paper presents the design of a fingerprint and iris-based framework for CBE invigilation. The framework comprises of modules for CBE, e-invigilation and control. The CBE module comprises of a network backbone, a server and several workstations. The e-invigilation module is designed to use high definition and resolution iris scanners such as Iris Shield-USB MK and CMITech BMT series to capture the iris image of the candidates for processing while the control module will handle the tasks of fingerprint-based authentication of examinees, process monitoring and relaying of situation reports.

Pattern Recognition Approaches in Music Analytics

Volume 5 Issue 2 June - August 2018

Review Paper

Pattern Recognition Approaches in Music Analytics

Makarand Velankar*, Parag Arun Kulkarni**

* Assistant Professor, Department of Information Technology, MKSSS's Cummins College of Engineering and PhD Research Scholar PICT, SPPU Pune, Maharashtra, India.

** Founder, Chief Scientist and CEO, iknowlation Research Labs Pvt. Ltd., Pune, Maharashtra, India.

Velankar, M., and Kulkarni, P. A (2018). Pattern recognition approaches in music analytics. i-manager’s Journal on Pattern Recognition, 5(2), 37-46. https://doi.org/10.26634/jpr.5.2.14784

Abstract

Content based Music Information Retrieval (MIR) has been a study matter for MIR research group since the inception of the group. Different pattern recognition paradigms are used for the diverse application for content-based music information retrieval. Music is a multidimensional phenomenon posing severe investigation tasks. Diverse tasks such as automatic music transcription, music recommendation, style identification, music classification, emotion modeling etc. requires quantitative and qualitative analysis. In spite of noteworthy efforts, the conclusions revealed shows latency over correctness achieved in different tasks. This paper covers different feature learning techniques used for music data in conventional audio pattern in different digital signal processing domains. Considering the remarkable improvements in results for applications related to speech and image processing using deep learning approach, similar efforts are attempted in the domain of music data analytics. Deep learning applied for music analytics applications are covered along with music adversaries reported. Future directions in conventional and deep learning approach with evaluation criteria for pattern recognition approaches in music analytics are explored.

Fingerprint Based Driver's Identification System

Volume 5 Issue 2 June - August 2018

Research Paper

Fingerprint Based Driver's Identification System

Inalegwu O. C.*, Maliki D. **, Agajo J.***, Ajao L. A.****, Abu A. D. *****

*-**Research Scholar, Department of Computer Engineering, Federal University of Technology, Minna, Nigeria.

***Lecturer, Department of Computer Engineering, Federal University of Technology Minna, Nigeria.

****Research Associate, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria.

*****ICT Instructor, Air Force Military School, Jos, Nigeria.

Inalegwu, O. C., Maliki, D., Agajo, J., Ajao, L. A., and Abu, A. D (2018). Fingerprint Based Driver's Identification System. i-manager’s Journal on Pattern Recognition, 5(2),30-36. https://doi.org/10.26634/jpr.5.2.15730

Abstract

This design work presents a proposed replacement to the current system used by the Federal Road Safety Commission (FRSC) for checking licensed/unlicensed drivers. It gives a faster and less tedious way of identifying registered and licensed road users using biometric captures. The system employs the use of an Arduino board to control and process the functioning of other peripherals: the fingerprint scanner and the Organic Light Emitting Diode (OLED) screen connected to it to achieve its purpose. The prototype system developed was able to displays driver's information on the OLED screen (Age, Name, Sex and License ID); the average response time of the system was also calculated to be 1.41 seconds, which is a good response time considering the system in question. The tested false acceptance rate and false rejection rates were relatively low (after a sample test with 25 individuals); at 4% and 8% respectively. Also, for its implementation, the components are readily available, relatively cheap and the system is one that can be easily adopted by the FRSC if access to their already existing database is granted. Consequently, it is safe to say that the developed system measured up to the design expectations; it meets the aim of a proposed replacement for the present analogue and easy to beat system employed by the FRSC.

Application of Machine Learning to the Diagnosis of Lower Respiratory Tract Infections in Paediatric Patients

Volume 5 Issue 2 June - August 2018

Research Paper

Application of Machine Learning to the Diagnosis of Lower Respiratory Tract Infections in Paediatric Patients

Olufunke C. Olayemi*, Adewale O. Sunday **, Olayemi O. Olasehinde***, Bolanle A. Ojokoh****, Adebayo O. Adetunmbi*****

*Lecturer, Department of Computer Science, Joseph Ayo Babalola University Ikeji-Arakeji, Osun State, Nigeria.

**,*****Professor, Department of Computer Science, School of Computing, Federal University of Technology (FUTA), Akure, Nigeria.

***Lecturer, Department of Computer Science, Federal Polytechnic Ile-Oluji, Ondo State, Nigeria.

****Associate Professor, Department of Computer Science, Federal University of Technology (FUTA), Akure, Nigeria.

Olayemi, O. C., Adewale, O. S., Olasehinde, O. O., Ojokoh, B. A., and Adetunmbi, A. O. (2018). Application of Machine Learning to the Diagnosis of Lower Respiratory Tract Infections in Paediatric Patients. i-manager’s Journal on Pattern Recognition, 5(2), 21-29. https://doi.org/10.26634/jpr.5.2.15538

Abstract

Lower Respiratory Tract Infection (LRTI) is a common infection among children in both tropical and subtropical regions which includes Africa, America and Asia. World Health Organization reported more than 2.5 million of deaths as a result of LRTI in 2012, late and untimely diagnosis of this infection is one of the factors responsible for its high mortality rate. This paper employed the use of machine learning techniques to diagnose the presence of LRTI in infants. The LRTI dataset obtained from Federal Medical centre (FMC) Owo in Ondo State was preprocessed and relevant attributes obtained from it as well as the whole preprocessed dataset were used to implement a Naïve bayes and K- nearest neighbor machine learning models using java programming language. The performance of the models were evaluated based on accuracy, sensitivity, specificity and precision. The result of Naïve bayes and k-nearest neighbour with all features (18) used shows 94.25% and 94.43% respectively. Naïve Bayes with information- based feature selection method shows accuracy of 99.60% while k-nearest neighbour shows 94.35% with 10 features. Also, Naïve Bayes with Correlation-based feature selection method shows accuracy of 95.40% while k-nearest neighbour shows 95.40% too with just six (6) features. The comparative results shows that Naïve bayes with information- based feature selection method performs stronger and better than others.

A Fuzzy Based Method for Diagnosis of Acne Skin Disease Severity

Volume 5 Issue 2 June - August 2018

Research Paper

A Fuzzy Based Method for Diagnosis of Acne Skin Disease Severity

Femi Emmanuel Ayo*, Joseph Bamidele Awotunde**, Sakinat Oluwabukonla Folorunso***, Ogundokun Roseline Oluwaseun****, P. S. Idoko*****, Jimoh Isiaka Adekunle******, Oladipo Idowu Dauda*******

*Assistant Lecturer, Department of Physical and Computer Sciences, McPherson University, Seriki Sotayo, Nigeria.

**Senior Tutor, Department of Computer Science, University of Ilorin, Ilorin, Nigeria.

***Lecturer II, Department of Mathematical Sciences, Olabisi Onabanjo University, Ago Iwoye, Nigeria.

****Lecturer, Department of Computer Science, Landmark University, Omu-Aran, Nigeria.

*****Principal Lecturer, Department of Computer Science, Kogi State Polytechnic, Lokoja, Nigeria.

******Principal Lecturer, Department of Computer Science, Federal Polytechnic, Offa, Nigeria.

*******Department of Computer Science, University of Ilorin, Ilorin, Nigeria.

Ayo, F. E., Awotunde, J. B., Folorunso, S. O., Oluwaseun, O. R., Idoko, P. S., and Adekunl, J. I. (2018). A Fuzzy Based Method for Diagnosis of Acne Skin Disease Severity. i-manager’s Journal on Pattern Recognition, 5(2), 10-20. https://doi.org/10.26634/jpr.5.2.15537

Abstract

Skin diseases are conditions that irritate, or affect the skin causing huge impact on a person’s day-to-day life. The tight schedule of people has greatly affected their availability to routine check-ups, thus keeping them away from visiting a doctor. The reputation of web-based medical systems is slowly becoming a paradigm to help people know the severity level of a disease. Acne skin disease ranks among the most popular skin disease and upsets the sebaceous glands, hence, routine check-ups could help prevent burns. In this paper, fuzzy based method is proposed for the identification of acne skin disease. This method is proposed to overcome the shortcoming of expert systems in previous methods. From literature, expert system reasoning is associated with uncertainty. Our proposed expert system uses fuzzy rules to resolve imprecision in the expert system reasoning. According to the evaluation results from the confusion matrix, modeled for evaluating the performance of the proposed fuzzy expert scheme, it was established that the scheme got 82% accuracy, which is indicative of a good performance. The designed fuzzy expert system showed a high level of recommendation, treatment advice and suggests the severity of acne skin disease in the patient.

A Framework for Fingerprint Liveness Detection Using Support Vector Machine Optimized by Genetic Algorithm

Volume 5 Issue 2 June - August 2018

Research Paper

A Framework for Fingerprint Liveness Detection Using Support Vector Machine Optimized by Genetic Algorithm

Yusuf Ibrahim*, Muhammed B. Mu’azu**, Emmanuel A. Adedokun***, Yusuf A. Sha’aban****

*,**** Lecturer, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria.

** Professor and Head, Department of Computer Engineering, Ahmadu Bello University, Zaria, Nigeria.

*** Department of Electrical Engineering, Ahmadu Bello University, Zaria, Nigeria.

Ibrahim, Y., Mu’azu, M. B., Adedokun, E. A.,and Sha’aban, Y. A (2018). A Framework for Fingerprint Liveness Detection Using Support Vector Machine Optimized by Genetic Algorithm. i-manager’s Journal on Pattern Recognition, 5(2), 1-9. https://doi.org/10.26634/jpr.5.2.15536

Abstract

Fingerprints are widely and successfully been used in a number of applications as a preferred biometric for personal identifications. However, current fingerprint authentication systems are vulnerable to direct spoof attacks at the sensor level as fake fingerprints artificially made to replicate genuine ones are now made using common materials such as silicone, gelatin, playdoh etc. This paper therefore implements a software based deep machine learning framework for classifying fingerprints images presented to the system as either been live or fake. Since typical fingerprint images are noisy, some preprocessing on the images were first of all performed using a decision based adaptive median filtering algorithm for de-noising and min-max normalization for enhancement. Features were then extracted using pre- trained Deep Convolutional Neural Network (DCNN) and their dimensionality reduced using Principal Component Analysis (PCA). Resulting features were then used to train a Support Vector Machine with Gaussian kernel optimized by Genetic Algorithm. The developed GA-SVM method was evaluated on the 3993 Biometrika datasets from the LivDet2009 database. The results obtained demonstrate robustness and effectiveness of the developed method in achieving good average liveness classification accuracy.

Vision Based Counterfeit Currency Detection System

Volume 5 Issue 1 March - May 2018

Research Paper

Vision Based Counterfeit Currency Detection System

Nidhi A.*, Soumak Chongder**, T. Shreekanth***

*-** BE Graduate, Department of Electronics and Communication Engineering, Sri Jayachamarajendra College of Engineering, Mysore, Karnataka, India.

*** Assistant Professor, Department of Electronics and Communication Engineering, Sri Jayachamarajendra College of Engineering, Mysore, Karnataka, India.

Nidhi, A., Chongder, S., and Shreekanth, T. (2018). Vision Based Counterfeit Currency Detection System. I-Manager’s Journal On Pattern Recognition, 5(1), 32-41. https://doi.org/10.26634/jpr.5.1.14298

Abstract

Counterfeit currency has always been a threat to the economy of the country. Despite the addition of several security features to prevent this, people have always found ways to duplicate currency. It is also difficult in most cases for a common man to identify a fake note and thus, one falls prey to such tricks of counterfeit currency. Hence, it becomes important to adopt newer and better methods to counter the same for all the denominations of currency in circulation, the most vulnerable being the ones with higher denomination. While there exist a number of ways to check for the correctness of the older variants of Indian currency, the newer lot that has been released however has incorporated in themselves, a number of changes and newer features. This paper aims to bring out some of the vital parameters to be extracted to distinguish between real and fake notes in the new set of Indian currency of denominations 2000, 500, 200, 50 and 10 brought out by the Reserve Bank of India (RBI), using various image processing techniques.

Pattern Recognition System For Condition Monitoring Of Overhead Power Distribution Line Insulators Using Curvelet and Contourlet Features

Volume 5 Issue 1 March - May 2018

Research Article

Pattern Recognition System For Condition Monitoring Of Overhead Power Distribution Line Insulators Using Curvelet and Contourlet Features

Potnuru Surya Prasad*, Bhima Prabhakara Rao**

*Ph.D Scholar, Jawaharlal Nehru Technological University, Kakinada, Andhra Pradesh, India.

**Programme Director, School of Nanotechnology, JNTU, Kakinada, Andhra Pradesh, India.

Prasad, P.S., and Rao, B.P., (2018). Pattern Recognition System For Condition Monitoring Of Overhead Power Distribution Line Insulators Using Curvelet And Contourlet Features. i-manager’s Journal on Pattern Recognition, 5(1), 21-31. https://doi.org/10.26634/jpr.5.1.14792

Abstract

The power distribution system is considered as the important component of a power system because the consistent delivery of power to the consumers depends on it. Due to massive growth in the consumption of power, the damaged insulators on the electric poles prompt the breakage of the power supply which leads to considerable loss occurring for the power industry and hence to the national economy. As the insulators protect the power distribution system from heavy transients, there must be a monitoring system to regularly check the condition of the insulators. Regular monitoring of the overhead power line insulators requires taking pictures of the poles, sending them to the processing unit and applying image processing techniques to classify the insulator health condition into either healthy or risky and subsequent necessary replacement of the damaged insulator can be done by the maintenance personnel. Using the above procedure, the breakage condition of the insulators can be determined. The insulator images are extracted from the acquired pole image input and then individual insulator's statistical features are obtained based on curvelet transform and contourlet transform coefficients. The obtained features of insulator images are given to SVM (Support Vector Machines) classifier in determining the health condition of an insulator and the experiment results are validated. The health condition monitoring of power system insulators can be done reliably and hence this method of automatic classification would reduce the human efforts to a greater extent.

Offline Signature Authentication System Using Machine Learning and Android Interface

Volume 5 Issue 1 March - May 2018

Research Paper

Offline Signature Authentication System Using Machine Learning and Android Interface

Nirmita Nagaraj*, Kiran Y.C**

*PG Scholar, Department of Computer Science and Engineering, B.N.M Institute of Technology, Bengaluru, Karnataka, India.

**Professor, Department of Computer Science and Engineering, B.N.M Institute of Technology, Bengaluru, Karnataka, India.

Nirmita, N., and Kiran, Y. C., (2018). Offline Signature Authentication System Using Machine Learning And Android Interface. i-manager’s Journal on Pattern Recognition, 5(1), 15-20. https://doi.org/10.26634/jpr.5.1.14583

Abstract

Signature is being widely used as a personal identification or a verification system, which also comes with wide variety of problems which is getting exposed to forgery. Human errors could add more complexity into the process, hence there is always a need for automated system. Verification can be either online or offline-based. Verification can be performed or accomplished in either ways i.e. online-based or offline-based. The online-based works on image which is digitally acquired as signature uses dynamic information of the signature, when the signature is signed. This paper proposes an offline-system which integrates Android, Matlab, Java where the whole algorithm or the heart of the process takes place in Matlab, Java provides the server and android acts as UI interface. For verification, techniques which are based on geometric features and corner features combined with the training of neural network have been used. Several geometric features have been combined which includes Occupancy region, region where Centroid exist, deviation, even pixels Harris and Scale Invariant Feature Transform (SIFT) features and also Kurtosis, Skewness. The proposed methodology technique includes pre-processing of a scanned signature image at the beginning. Neural network is used as a decision maker for real or forged, while the efficiency of correct recognition is around 90.24% with a threshold of genuine at 60%. The simulation shows that the proposed method has a clear discriminative nature between real and forged signatures.

An Idea To Design A Trapping System To Enhance Security

Volume 5 Issue 1 March - May 2018

Research Paper

An Idea To Design A Trapping System To Enhance Security

Umesh Kumar Pandey*, Snehlata Barde**

*Assistant Professor (Contract Based), Department Of Computer Science, Rajeev Gandhi Govt. Post-Graduate Autonomous College, Ambikapur, Chhattisgarh, India.

**Associate Professor, MATS University, Raipur, Chhattisgarh, India.

Pandey, U.K., and Barde, S. (2018).An Idea To Design A Trapping System To Enhance Security . i-manager’s Journal on Pattern Recognition, 5(1), 10-14. https://doi.org/10.26634/jpr.5.1.14793

Abstract

In today’s world, it is known that, it is a big problem to search for a person who is lost, who has stolen any items from us, who is accused of any crime or who performs any unauthorized work in an organization. Sometimes, we also know about this person, but we are unable to find the current or past working details, acting, or moving status of this accused person; how will we find that person, secure society and aware a specific area of people from human crimes is a big challenge of the security system. For this, an effort has been made to design an online trapping system based on web enable application to solve the above problems. Data were gathered from people who wants to complain online, control all services, and trap the criminal. This system provides the facility to trap a person who is accused in any crime and moving anywhere.