Matthew Dailey - Academia.edu (original) (raw)

Papers by Matthew Dailey

Research paper thumbnail of Evaluating the Effectiveness of Sentiment-Based Models for Stock Price Prediction

Research paper thumbnail of Multimodal Sentiment Analysis on Video Streams using Lightweight Deep Neural Networks

Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods, 2021

Research paper thumbnail of A text-graphic character Captcha for password identification

2004 IEEE Region 10 Conference TENCON 2004.

We propose a new construct, the Text-Graphics Character (TGC) CAPTCHA, for preventing dictionary ... more We propose a new construct, the Text-Graphics Character (TGC) CAPTCHA, for preventing dictionary attacks against password authenticated systems allowing remote access via dumb terminals. Password authentication is commonly used for computer access control. But password authenticated systems are prone to dictionary attacks, in which attackers repeatedly attempt to gain access using the entries in a list of frequentlyused passwords. CAPTCHAs (Completely Automated Public Turing tests to tell Computers and Humans Apart) are currently being used to prevent automated "bots" from registering for email accounts. They have also been suggested as a means for preventing dictionary attacks. However, current CAPTCHAs are unsuitable for text-based remote access. Our TGC CAPTCHA fills this gap. In this paper, we define the TGC CAPTCHA, prove that it is a (secure) CAPTCHA, demonstrate its utility in a prototype based on the SSH (Secure Shell) protocol suite, and provide empirical evidence that the test is easy for humans and hard for machines. We believe that the system will not only help improve the security of servers allowing remote terminal access, but also encourage a healthy spirit of competition in the fields of pattern recognition, computer graphics, and psychology.

Research paper thumbnail of Fuzzy-C-Mean Determines the Principle Component Pairs to Estimate the Degree of Emotion from Facial Expressions

Lecture Notes in Computer Science, 2005

Although many systems exist for automatic classification of faces according to their emotional ex... more Although many systems exist for automatic classification of faces according to their emotional expression, these systems do not explicitly estimate the strength of given expressions. This paper describes and empirically evaluates an algorithm capable of estimating the degree to which a face expresses a given emotion. The system first aligns and normalizes an input face image, then applies a filter bank of Gabor wavelets and reduces the data's dimensionality via principal components analysis. Finally, an unsupervised Fuzzy-C-Mean clustering algorithm is employed recursively on the same set of data to find the best pair of principle components from the amount of alignment of the cluster centers on a straight line. The cluster memberships are then mapped to degrees of a facial expression (i.e. less Happy, moderately happy, and very happy). In a test on 54 previously unseen happy faces., we find an orderly mapping of faces to clusters as the subject's face moves from a neutral to very happy emotional display. Similar results are observed on 78 previously unseen surprised faces.

Research paper thumbnail of Empirical Study of Car License Plates Recognition

JUTI: Jurnal Ilmiah Teknologi Informasi, 2015

Jumlah kendaraan bermotor di jalan bertambah secara signifikan. Oleh karena itu, dibutuhkan penge... more Jumlah kendaraan bermotor di jalan bertambah secara signifikan. Oleh karena itu, dibutuhkan pengenal atau identitas dari tiap kendaraan yaitu menggunakan plat nomor. Dari plat nomor, dapat diketahui pemilik dan informasi mengenai kendaraan bermotor. Plat nomor juga dapat digunakan untuk mempermudah sistem manajemen lalu lintas. Sistem manajemen lalu lintas diterapkan di jalan raya untuk mengawasi kecepatan mobil yang melebihi batas kecepatan maksimum serta digunakan untuk pengamanan parkir. Pada studi ini dilakukan perbandingan dalam hal pendeteksian lokasi plat nomor kendaraan, hasil segmentasi plat dan pengekstraksian karakter yang terdapat pada plat nomor kendaraan. Algoritma yang dibandingkan untuk pendeteksian plat kendaraan yaitu pendekatan histogram, sliding concetric window, dan Haar-like wavelet cascade. Untuk algoritma segmentasi dibandingkan antara morphological, Otsu dan Laplacian. Sedangkan algoritma pengekstraksian karakter plat digunakan algoritma template matching dan neural network. Dari proses perbandingan metode tersebut dicari hasil pengenalan karakter plat nomor yang akurat dan metode yang efisien untuk dapat diterapkan. Kombinasi terbaik dengan akurasi tertinggi adalah Haar cascade-Laplacian-neural network.

Research paper thumbnail of A comparative study on Thai word segmentation approaches

2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, 2008

In this paper, we analyze and compare various approaches to Thai word segmentation. Word segmenta... more In this paper, we analyze and compare various approaches to Thai word segmentation. Word segmentation approaches can be classified into two distinct types: dictionarybased (DCB) and machine learning-based (MLB). DCB approaches rely on sets of stored terms to parse and segment input text. MLB approaches, on the other hand, rely on statistical models estimated from training corpora using machine learning techniques. We compare two algorithms based on the DCB approach: longest matching and maximal matching. We compare four algorithms based on the MLB approach: Naive Bayes (NB), decision tree, Support Vector Machine (SVM), and Conditional Random Field (CRF). In a series of experiments, the DCB approach performed better than the NB, decision tree and SVM algorithms from the MLB approach. However, the best performing algorithm was the CRF algorithm, with precision and recall of 95.79% and 94.98%, respectively. We believe that the CRF is the best existing statistical model for problems like Thai word segmentation.

Research paper thumbnail of Towards Knowledge-Based Affective Interaction: Situational Interpretation of Affect

Lecture Notes in Computer Science

Human-to-computer interaction in a variety of applications could benefit if systems could accurat... more Human-to-computer interaction in a variety of applications could benefit if systems could accurately analyze and respond to their users' affect. Although a great deal of research has been conducted on affect recognition, very little of this work has considered what is the appropriate information to extract in specific situations. Towards understanding how specific applications such as affective tutoring and affective entertainment could benefit, we present two experiments. In the first experiment, we found that students' facial expressions, together with their body actions, gave little information about their internal emotion per se but they would be useful features for predicting their self-reported "true" mental state. In the second experiment, we found significant differences between the facial expressions and self-reported affective state of viewers watching a movie sequence. Our results suggest that the noisy relationship between observable gestures and underlying affect must be accounted for when designing affective computing applications.

Research paper thumbnail of SLA-driven adaptive resource management for Web applications on a heterogeneous compute cloud

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2009

Current service-level agreements (SLAs) offered by cloud providers make guarantees about quality ... more Current service-level agreements (SLAs) offered by cloud providers make guarantees about quality attributes such as availability. However, although one of the most important quality attributes from the perspective of the users of a cloud-based Web application is its response time, current SLAs do not guarantee response time. Satisfying a maximum average response time guarantee for Web applications is difficult due to unpredictable traffic patterns, but in this paper we show how it can be accomplished through dynamic resource allocation in a virtual Web farm. We present the design and implementation of a working prototype built on a EUCALYPTUS-based heterogeneous compute cloud that actively monitors the response time of each virtual machine assigned to the farm and adaptively scales up the application to satisfy a SLA promising a specific average response time. We demonstrate the feasibility of the approach in an experimental evaluation with a testbed cloud and a synthetic workload. Adaptive resource management has the potential to increase the usability of Web applications while maximizing resource utilization.

Research paper thumbnail of SLA-driven automatic bottleneck detection and resolution for read intensive multi-tier applications hosted on a cloud

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2010

A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers ... more A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers of services. However, the current SLAs offered by cloud providers do not address response time, which, from the user's point of view, is the most important quality attribute for Web applications. Satisfying a maximum average response time guarantee for Web applications is difficult for two main reasons: first, traffic patterns are unpredictable; second, the complex nature of multi-tier Web applications increases the difficulty of identifying bottlenecks and resolving them automatically. This paper presents a working prototype system that automatically detects and resolves bottlenecks in a multi-tier Web application hosted on a EUCALYPTUS-based cloud in order to satisfy specific maximum response time requirements. We demonstrate the feasibility of the approach in an experimental evaluation with a testbed cloud and a synthetic workload. Automatic bottleneck detection and resolution under dynamic resource management has the potential to enable cloud providers to provide SLAs for Web applications that guarantee specific response time requirements.

Research paper thumbnail of Segmentation of text and non-text in on-line handwritten patient record based on spatio-temporal analysis

Research paper thumbnail of Providing objective feedback on skill assessment in a dental surgical training simulator

Research paper thumbnail of Steganography in Thai text

2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, 2008

Research paper thumbnail of Rapid Detection of Many Object Instances

Lecture Notes in Computer Science, 2009

Research paper thumbnail of Analysis Performance of Fast Image Encryption

JUTI: Jurnal Ilmiah Teknologi Informasi, 2014

Perkembangan teknologi mengakibatkan peningkatan kebutuhan pengiriman data melalui media internet... more Perkembangan teknologi mengakibatkan peningkatan kebutuhan pengiriman data melalui media internet. Banyak pengiriman data yang membutuhkan keamanan dalam pengirimannya untuk berbagai keperluan. Enkripsi data merupakan salah satu topic pengamanan yang banyak dilakukan penelitian dengan tujuan untuk mengamankan data yang dikirimkan melalui media internet. Salah satu data yang banyak digunakan adalah data citra. Citra merupakan data yang memiliki kapasitas besar dan memiliki sifat perulangan yang tinggi sehingga dibutuhkan metode tertentu untuk melakukan proses enkripsi dan dekripsi citra. Permutasi dan difusi merupakan cara yang banyak digunakan untuk melakukan enkripsi citra. Permutasi bertujuan untuk mengacak posisi citra sedangkan difusi merubah nilai citra. Permutasi dan difusi banyak dilakukan sebagai dua tahap yang berbeda sehingga dibutuhkan dua kali pembacaan citra. Sebuah algoritma untuk menggabungkan proses permutasi dan difusi sehingga hanya diperlukan satu kali pembacaan citra untuk melakukan enkripsi telah diajukan. Selain permutasi dan difusi, fungsi chaos juga digunakan dalam algoritma tersebut karena kemampuannya untuk menghasilkan angka random yang sangat sensitif terhadap beberapa parameter. Dengan ide demikian, algoritma akan cepat untuk melakukan proses enkripsi dan dekripsi. Dalam penelitian ini dianalisis kinerja algoritma gabungan permutasi dan difusi menggunakan fungsi chaos. Analisis dilakukan dengan mengimplementasikan algoritma, mendapatkan waktu yang dibutuhkan untuk proses enkripsi dan dekripsi serta membandingkannnya dengan algoritma baku yang telah banyak digunakan, Advanced Encryption Standart (AES).

Research paper thumbnail of Discovering repetitive patterns in facade images using a RANSAC-style algorithm

ISPRS Journal of Photogrammetry and Remote Sensing, 2014

Research paper thumbnail of Anatomical sketch understanding: Recognizing explicit and implicit structure

Artificial Intelligence in Medicine, 2007

Objective: Sketching is ubiquitous in medicine. Physicians commonly use sketches as part of their... more Objective: Sketching is ubiquitous in medicine. Physicians commonly use sketches as part of their note taking in patient records and to help convey diagnoses and treatments to patients. Medical students frequently use sketches to help them think through clinical problems in individual and group problem solving. Applications ranging from automated patient records to medical education software could benefit greatly from the richer and more natural interfaces that would be enabled by the ability to understand sketches. In this paper we take the first steps toward developing a system that can understand anatomical sketches. Methods: Understanding an anatomical sketch requires the ability to recognize what anatomical structure has been sketched and from what view (e.g. parietal view of the brain), as well as to identify the anatomical parts and their locations in the sketch (e.g. parts of the brain), even if they have not been explicitly drawn. We present novel algorithms for sketch recognition and for part identification. We evaluate the accuracy of the recognition algorithm on sketches obtained from medical students. We evaluate the part identification algorithm by comparing its results to the judgment of an experienced physician. Results: The sketch recognition algorithm achieves a recognition accuracy of 75.5%, far above the baseline random classification accuracy of 6.7%. Comparison of the results of the part identification algorithm with the judgment of an experienced § This is a revised and extended version of the paper, ''Anatomical sketch under standing: Recognizing explicit and implicit structure,'' by P. Haddawy, M.N. Dailey, P. Kaewruen, and N. Sarakhette, presented in the 10th Conference on Artificial Intelligence in Medicine (AIME 05).

Research paper thumbnail of Slam with KLT Point Features

cs.ait.ac.th

In the simultaneous localization and mapping (SLAM) prob-lem, a mobile robot must localize itself... more In the simultaneous localization and mapping (SLAM) prob-lem, a mobile robot must localize itself in an unknown en-vironment using its sensors and at the same time construct a map of that environment. While SLAM utilizing costly (expensive, heavy and slow) laser range ...

Research paper thumbnail of Interactive Segmentation and Three-Dimension Reconstruction for Cone-Beam Computed-Tomography Images

submitted to The NECTEC …, 2008

We develop a novel interactive segmentation and 3-D visualization system for cone-beam computed-t... more We develop a novel interactive segmentation and 3-D visualization system for cone-beam computed-tomography (CBCT) data in a personal computer environment. Its design includes both a new user interface to ease the interactive manual segmentation of the ...

Research paper thumbnail of Clustering Human Behaviors With Dynamic Time Warping and Hidden Markov Models for a Video Surveillance System

cs.ait.ac.th

We propose and experimentally evaluate a new method for clustering human behaviors that is suitab... more We propose and experimentally evaluate a new method for clustering human behaviors that is suitable for bootstrapping an anomaly detection module for intelligent video surveillance systems. The method uses dynamic time warping, agglomerative hierarchical clustering, and hidden Markov models to provide an initial partitioning of a set of observation sequences then automatically identifies where to cut off the hierarchical clustering dendrogram. We show that the method is extremely effective, providing 100% accuracy in separating anomalous from typical behaviors on real-world testbed video surveillance data.

Research paper thumbnail of Evaluating the Effectiveness of Sentiment-Based Models for Stock Price Prediction

Research paper thumbnail of Multimodal Sentiment Analysis on Video Streams using Lightweight Deep Neural Networks

Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods, 2021

Research paper thumbnail of A text-graphic character Captcha for password identification

2004 IEEE Region 10 Conference TENCON 2004.

We propose a new construct, the Text-Graphics Character (TGC) CAPTCHA, for preventing dictionary ... more We propose a new construct, the Text-Graphics Character (TGC) CAPTCHA, for preventing dictionary attacks against password authenticated systems allowing remote access via dumb terminals. Password authentication is commonly used for computer access control. But password authenticated systems are prone to dictionary attacks, in which attackers repeatedly attempt to gain access using the entries in a list of frequentlyused passwords. CAPTCHAs (Completely Automated Public Turing tests to tell Computers and Humans Apart) are currently being used to prevent automated "bots" from registering for email accounts. They have also been suggested as a means for preventing dictionary attacks. However, current CAPTCHAs are unsuitable for text-based remote access. Our TGC CAPTCHA fills this gap. In this paper, we define the TGC CAPTCHA, prove that it is a (secure) CAPTCHA, demonstrate its utility in a prototype based on the SSH (Secure Shell) protocol suite, and provide empirical evidence that the test is easy for humans and hard for machines. We believe that the system will not only help improve the security of servers allowing remote terminal access, but also encourage a healthy spirit of competition in the fields of pattern recognition, computer graphics, and psychology.

Research paper thumbnail of Fuzzy-C-Mean Determines the Principle Component Pairs to Estimate the Degree of Emotion from Facial Expressions

Lecture Notes in Computer Science, 2005

Although many systems exist for automatic classification of faces according to their emotional ex... more Although many systems exist for automatic classification of faces according to their emotional expression, these systems do not explicitly estimate the strength of given expressions. This paper describes and empirically evaluates an algorithm capable of estimating the degree to which a face expresses a given emotion. The system first aligns and normalizes an input face image, then applies a filter bank of Gabor wavelets and reduces the data's dimensionality via principal components analysis. Finally, an unsupervised Fuzzy-C-Mean clustering algorithm is employed recursively on the same set of data to find the best pair of principle components from the amount of alignment of the cluster centers on a straight line. The cluster memberships are then mapped to degrees of a facial expression (i.e. less Happy, moderately happy, and very happy). In a test on 54 previously unseen happy faces., we find an orderly mapping of faces to clusters as the subject's face moves from a neutral to very happy emotional display. Similar results are observed on 78 previously unseen surprised faces.

Research paper thumbnail of Empirical Study of Car License Plates Recognition

JUTI: Jurnal Ilmiah Teknologi Informasi, 2015

Jumlah kendaraan bermotor di jalan bertambah secara signifikan. Oleh karena itu, dibutuhkan penge... more Jumlah kendaraan bermotor di jalan bertambah secara signifikan. Oleh karena itu, dibutuhkan pengenal atau identitas dari tiap kendaraan yaitu menggunakan plat nomor. Dari plat nomor, dapat diketahui pemilik dan informasi mengenai kendaraan bermotor. Plat nomor juga dapat digunakan untuk mempermudah sistem manajemen lalu lintas. Sistem manajemen lalu lintas diterapkan di jalan raya untuk mengawasi kecepatan mobil yang melebihi batas kecepatan maksimum serta digunakan untuk pengamanan parkir. Pada studi ini dilakukan perbandingan dalam hal pendeteksian lokasi plat nomor kendaraan, hasil segmentasi plat dan pengekstraksian karakter yang terdapat pada plat nomor kendaraan. Algoritma yang dibandingkan untuk pendeteksian plat kendaraan yaitu pendekatan histogram, sliding concetric window, dan Haar-like wavelet cascade. Untuk algoritma segmentasi dibandingkan antara morphological, Otsu dan Laplacian. Sedangkan algoritma pengekstraksian karakter plat digunakan algoritma template matching dan neural network. Dari proses perbandingan metode tersebut dicari hasil pengenalan karakter plat nomor yang akurat dan metode yang efisien untuk dapat diterapkan. Kombinasi terbaik dengan akurasi tertinggi adalah Haar cascade-Laplacian-neural network.

Research paper thumbnail of A comparative study on Thai word segmentation approaches

2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, 2008

In this paper, we analyze and compare various approaches to Thai word segmentation. Word segmenta... more In this paper, we analyze and compare various approaches to Thai word segmentation. Word segmentation approaches can be classified into two distinct types: dictionarybased (DCB) and machine learning-based (MLB). DCB approaches rely on sets of stored terms to parse and segment input text. MLB approaches, on the other hand, rely on statistical models estimated from training corpora using machine learning techniques. We compare two algorithms based on the DCB approach: longest matching and maximal matching. We compare four algorithms based on the MLB approach: Naive Bayes (NB), decision tree, Support Vector Machine (SVM), and Conditional Random Field (CRF). In a series of experiments, the DCB approach performed better than the NB, decision tree and SVM algorithms from the MLB approach. However, the best performing algorithm was the CRF algorithm, with precision and recall of 95.79% and 94.98%, respectively. We believe that the CRF is the best existing statistical model for problems like Thai word segmentation.

Research paper thumbnail of Towards Knowledge-Based Affective Interaction: Situational Interpretation of Affect

Lecture Notes in Computer Science

Human-to-computer interaction in a variety of applications could benefit if systems could accurat... more Human-to-computer interaction in a variety of applications could benefit if systems could accurately analyze and respond to their users' affect. Although a great deal of research has been conducted on affect recognition, very little of this work has considered what is the appropriate information to extract in specific situations. Towards understanding how specific applications such as affective tutoring and affective entertainment could benefit, we present two experiments. In the first experiment, we found that students' facial expressions, together with their body actions, gave little information about their internal emotion per se but they would be useful features for predicting their self-reported "true" mental state. In the second experiment, we found significant differences between the facial expressions and self-reported affective state of viewers watching a movie sequence. Our results suggest that the noisy relationship between observable gestures and underlying affect must be accounted for when designing affective computing applications.

Research paper thumbnail of SLA-driven adaptive resource management for Web applications on a heterogeneous compute cloud

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2009

Current service-level agreements (SLAs) offered by cloud providers make guarantees about quality ... more Current service-level agreements (SLAs) offered by cloud providers make guarantees about quality attributes such as availability. However, although one of the most important quality attributes from the perspective of the users of a cloud-based Web application is its response time, current SLAs do not guarantee response time. Satisfying a maximum average response time guarantee for Web applications is difficult due to unpredictable traffic patterns, but in this paper we show how it can be accomplished through dynamic resource allocation in a virtual Web farm. We present the design and implementation of a working prototype built on a EUCALYPTUS-based heterogeneous compute cloud that actively monitors the response time of each virtual machine assigned to the farm and adaptively scales up the application to satisfy a SLA promising a specific average response time. We demonstrate the feasibility of the approach in an experimental evaluation with a testbed cloud and a synthetic workload. Adaptive resource management has the potential to increase the usability of Web applications while maximizing resource utilization.

Research paper thumbnail of SLA-driven automatic bottleneck detection and resolution for read intensive multi-tier applications hosted on a cloud

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2010

A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers ... more A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers of services. However, the current SLAs offered by cloud providers do not address response time, which, from the user's point of view, is the most important quality attribute for Web applications. Satisfying a maximum average response time guarantee for Web applications is difficult for two main reasons: first, traffic patterns are unpredictable; second, the complex nature of multi-tier Web applications increases the difficulty of identifying bottlenecks and resolving them automatically. This paper presents a working prototype system that automatically detects and resolves bottlenecks in a multi-tier Web application hosted on a EUCALYPTUS-based cloud in order to satisfy specific maximum response time requirements. We demonstrate the feasibility of the approach in an experimental evaluation with a testbed cloud and a synthetic workload. Automatic bottleneck detection and resolution under dynamic resource management has the potential to enable cloud providers to provide SLAs for Web applications that guarantee specific response time requirements.

Research paper thumbnail of Segmentation of text and non-text in on-line handwritten patient record based on spatio-temporal analysis

Research paper thumbnail of Providing objective feedback on skill assessment in a dental surgical training simulator

Research paper thumbnail of Steganography in Thai text

2008 5th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, 2008

Research paper thumbnail of Rapid Detection of Many Object Instances

Lecture Notes in Computer Science, 2009

Research paper thumbnail of Analysis Performance of Fast Image Encryption

JUTI: Jurnal Ilmiah Teknologi Informasi, 2014

Perkembangan teknologi mengakibatkan peningkatan kebutuhan pengiriman data melalui media internet... more Perkembangan teknologi mengakibatkan peningkatan kebutuhan pengiriman data melalui media internet. Banyak pengiriman data yang membutuhkan keamanan dalam pengirimannya untuk berbagai keperluan. Enkripsi data merupakan salah satu topic pengamanan yang banyak dilakukan penelitian dengan tujuan untuk mengamankan data yang dikirimkan melalui media internet. Salah satu data yang banyak digunakan adalah data citra. Citra merupakan data yang memiliki kapasitas besar dan memiliki sifat perulangan yang tinggi sehingga dibutuhkan metode tertentu untuk melakukan proses enkripsi dan dekripsi citra. Permutasi dan difusi merupakan cara yang banyak digunakan untuk melakukan enkripsi citra. Permutasi bertujuan untuk mengacak posisi citra sedangkan difusi merubah nilai citra. Permutasi dan difusi banyak dilakukan sebagai dua tahap yang berbeda sehingga dibutuhkan dua kali pembacaan citra. Sebuah algoritma untuk menggabungkan proses permutasi dan difusi sehingga hanya diperlukan satu kali pembacaan citra untuk melakukan enkripsi telah diajukan. Selain permutasi dan difusi, fungsi chaos juga digunakan dalam algoritma tersebut karena kemampuannya untuk menghasilkan angka random yang sangat sensitif terhadap beberapa parameter. Dengan ide demikian, algoritma akan cepat untuk melakukan proses enkripsi dan dekripsi. Dalam penelitian ini dianalisis kinerja algoritma gabungan permutasi dan difusi menggunakan fungsi chaos. Analisis dilakukan dengan mengimplementasikan algoritma, mendapatkan waktu yang dibutuhkan untuk proses enkripsi dan dekripsi serta membandingkannnya dengan algoritma baku yang telah banyak digunakan, Advanced Encryption Standart (AES).

Research paper thumbnail of Discovering repetitive patterns in facade images using a RANSAC-style algorithm

ISPRS Journal of Photogrammetry and Remote Sensing, 2014

Research paper thumbnail of Anatomical sketch understanding: Recognizing explicit and implicit structure

Artificial Intelligence in Medicine, 2007

Objective: Sketching is ubiquitous in medicine. Physicians commonly use sketches as part of their... more Objective: Sketching is ubiquitous in medicine. Physicians commonly use sketches as part of their note taking in patient records and to help convey diagnoses and treatments to patients. Medical students frequently use sketches to help them think through clinical problems in individual and group problem solving. Applications ranging from automated patient records to medical education software could benefit greatly from the richer and more natural interfaces that would be enabled by the ability to understand sketches. In this paper we take the first steps toward developing a system that can understand anatomical sketches. Methods: Understanding an anatomical sketch requires the ability to recognize what anatomical structure has been sketched and from what view (e.g. parietal view of the brain), as well as to identify the anatomical parts and their locations in the sketch (e.g. parts of the brain), even if they have not been explicitly drawn. We present novel algorithms for sketch recognition and for part identification. We evaluate the accuracy of the recognition algorithm on sketches obtained from medical students. We evaluate the part identification algorithm by comparing its results to the judgment of an experienced physician. Results: The sketch recognition algorithm achieves a recognition accuracy of 75.5%, far above the baseline random classification accuracy of 6.7%. Comparison of the results of the part identification algorithm with the judgment of an experienced § This is a revised and extended version of the paper, ''Anatomical sketch under standing: Recognizing explicit and implicit structure,'' by P. Haddawy, M.N. Dailey, P. Kaewruen, and N. Sarakhette, presented in the 10th Conference on Artificial Intelligence in Medicine (AIME 05).

Research paper thumbnail of Slam with KLT Point Features

cs.ait.ac.th

In the simultaneous localization and mapping (SLAM) prob-lem, a mobile robot must localize itself... more In the simultaneous localization and mapping (SLAM) prob-lem, a mobile robot must localize itself in an unknown en-vironment using its sensors and at the same time construct a map of that environment. While SLAM utilizing costly (expensive, heavy and slow) laser range ...

Research paper thumbnail of Interactive Segmentation and Three-Dimension Reconstruction for Cone-Beam Computed-Tomography Images

submitted to The NECTEC …, 2008

We develop a novel interactive segmentation and 3-D visualization system for cone-beam computed-t... more We develop a novel interactive segmentation and 3-D visualization system for cone-beam computed-tomography (CBCT) data in a personal computer environment. Its design includes both a new user interface to ease the interactive manual segmentation of the ...

Research paper thumbnail of Clustering Human Behaviors With Dynamic Time Warping and Hidden Markov Models for a Video Surveillance System

cs.ait.ac.th

We propose and experimentally evaluate a new method for clustering human behaviors that is suitab... more We propose and experimentally evaluate a new method for clustering human behaviors that is suitable for bootstrapping an anomaly detection module for intelligent video surveillance systems. The method uses dynamic time warping, agglomerative hierarchical clustering, and hidden Markov models to provide an initial partitioning of a set of observation sequences then automatically identifies where to cut off the hierarchical clustering dendrogram. We show that the method is extremely effective, providing 100% accuracy in separating anomalous from typical behaviors on real-world testbed video surveillance data.