O câncer de pele é a neoplasia maligna mais incidente, no Brasil correspondendo a aproximadamente 29% de todos os tumores diagnosticados. O melanoma maligno é sua apresentação menos frequente, com cerca de 1% de todos os diagnósticos, porém com maior índice de mortalidade, comparado com outros tumores não melanomas. O fato de o melanoma ser considerado um tipo de câncer muito agressivo e não ser radiossensível, um diagnóstico precoce garante um alto índice de sobrevida ao paciente. O método mais utilizado para a análise de melanomas é a regra do ABCD, que consiste em extrair características de: assimetria, borda, coloração e diâmetro das lesões, a fim de diferenciá-las das demais lesões de pele. A dermatoscopia é uma técnica não invasiva, que permite fazer o reconhecimento das estruturas superficiais da pele com uma ampliação de até 400x, o que facilita a análise das lesões. A contribuição deste presente trabalho consiste na análise de lesões associada à técnica de processamento de imagem digital, com o objetivo de fornecer uma ferramenta de auxílio ao médico, dando um amparo no momento de se fornecer um diagnóstico ao paciente. O software desenvolvido neste trabalho utiliza imagens dermatoscópicas de lesões de pele associado à regra do ABCD, fornecendo por meio da análise destas características, um diagnóstico, se a lesão é melanoma ou não melanoma. / Skin cancer is the most frequent malignant tumor in Brazil representing approximately 29% of all diagnosed tumors. Malignant melanoma is the least common presentation, with about 1% of all diagnoses, but with higher mortality rate compared to other non-melanoma tumors. Early diagnosis ensures a high patient survival rate. The most used method for the analysis of these lesions is the ABCD rule, where features such as asymmetry, border, color and diameter of lesions are extracted. Dermatoscopy is a noninvasive technique that allows the recognition of skin structure with a magnification of up 400x. A good diagnosis is associated with the degree of subjectivity of the operator doctor. The contribution of this present work consists in the analysis of lesions associated with digital image processing technique, with the goal of provide a support tool to doctor, giving one support in time to give a diagnosis to the patient. The software developed here with dermatoscopic images of skin lesions associated with ABCD rule, giving a diagnosis, by the analysis of the lesion characteristics, indicating if the lesion is melanoma or no melanoma.

O câncer de pele é a neoplasia maligna mais incidente, no Brasil correspondendo a aproximadamente 29% de todos os tumores diagnosticados. O melanoma maligno é sua apresentação menos frequente, com cerca de 1% de todos os diagnósticos, porém com maior índice de mortalidade, comparado com outros tumores não melanomas. O fato de o melanoma ser considerado um tipo de câncer muito agressivo e não ser radiossensível, um diagnóstico precoce garante um alto índice de sobrevida ao paciente. O método mais utilizado para a análise de melanomas é a regra do ABCD, que consiste em extrair características de: assimetria, borda, coloração e diâmetro das lesões, a fim de diferenciá-las das demais lesões de pele. A dermatoscopia é uma técnica não invasiva, que permite fazer o reconhecimento das estruturas superficiais da pele com uma ampliação de até 400x, o que facilita a análise das lesões. A contribuição deste presente trabalho consiste na análise de lesões associada à técnica de processamento de imagem digital, com o objetivo de fornecer uma ferramenta de auxílio ao médico, dando um amparo no momento de se fornecer um diagnóstico ao paciente. O software desenvolvido neste trabalho utiliza imagens dermatoscópicas de lesões de pele associado à regra do ABCD, fornecendo por meio da análise destas características, um diagnóstico, se a lesão é melanoma ou não melanoma. / Skin cancer is the most frequent malignant tumor in Brazil representing approximately 29% of all diagnosed tumors. Malignant melanoma is the least common presentation, with about 1% of all diagnoses, but with higher mortality rate compared to other non-melanoma tumors. Early diagnosis ensures a high patient survival rate. The most used method for the analysis of these lesions is the ABCD rule, where features such as asymmetry, border, color and diameter of lesions are extracted. Dermatoscopy is a noninvasive technique that allows the recognition of skin structure with a magnification of up 400x. A good diagnosis is associated with the degree of subjectivity of the operator doctor. The contribution of this present work consists in the analysis of lesions associated with digital image processing technique, with the goal of provide a support tool to doctor, giving one support in time to give a diagnosis to the patient. The software developed here with dermatoscopic images of skin lesions associated with ABCD rule, giving a diagnosis, by the analysis of the lesion characteristics, indicating if the lesion is melanoma or no melanoma.

<p>Methods for assessment of body composition allow studies of the complex relationships between body composition and the wellbeing of humans. The amount and distribution of adipose tissue is of great importance in these studies. This thesis describes, proposes, and evaluates new methods for assessment of body composition (amount and distribution of adipose tissue) using magnetic resonance imaging (MRI). The thesis focuses on the use of MRI but includes results from computed tomography (CT) and dual energy X-ray absorptiometry (DEXA). The subject data used originates from cohorts recruited solely for the purpose of the included studies and from the “Prospective Investigation of Vasculature in Uppsala Seniors” (PIVUS) and the Sib-pair (within the framework of the Swedish Obese Subjects - SOS study) studies.</p><p>The included studies propose a new acquisition method for whole-body analysis of adipose tissue, a fully automated post processing algorithm for segmentation of visceral and subcutaneous adipose tissue from abdominal MRI data, and anthropometrical measurements for practical estimations of body composition.</p><p>The proposed acquisition method for whole-body adipose tissue analysis simplified the analysis of adipose tissue and the results strongly correlated with the results from CT and DEXA analysis. The fully automated post processing algorithm gave reproducible results with relatively high accuracy. Transverse and sagittal abdominal diameters gave information about subcutaneous and visceral adipose tissue, respectively, and an elliptical approximation was found useful in estimation of total amount of abdominal adipose tissue.</p><p>The methods proposed in this thesis were found useful for assessment of body composition. The methods were developed with clinical practice in mind and all proposed methods have been used in further studies for assessment of body composition.</p>

Methods for assessment of body composition allow studies of the complex relationships between body composition and the wellbeing of humans. The amount and distribution of adipose tissue is of great importance in these studies. This thesis describes, proposes, and evaluates new methods for assessment of body composition (amount and distribution of adipose tissue) using magnetic resonance imaging (MRI). The thesis focuses on the use of MRI but includes results from computed tomography (CT) and dual energy X-ray absorptiometry (DEXA). The subject data used originates from cohorts recruited solely for the purpose of the included studies and from the “Prospective Investigation of Vasculature in Uppsala Seniors” (PIVUS) and the Sib-pair (within the framework of the Swedish Obese Subjects - SOS study) studies. The included studies propose a new acquisition method for whole-body analysis of adipose tissue, a fully automated post processing algorithm for segmentation of visceral and subcutaneous adipose tissue from abdominal MRI data, and anthropometrical measurements for practical estimations of body composition. The proposed acquisition method for whole-body adipose tissue analysis simplified the analysis of adipose tissue and the results strongly correlated with the results from CT and DEXA analysis. The fully automated post processing algorithm gave reproducible results with relatively high accuracy. Transverse and sagittal abdominal diameters gave information about subcutaneous and visceral adipose tissue, respectively, and an elliptical approximation was found useful in estimation of total amount of abdominal adipose tissue. The methods proposed in this thesis were found useful for assessment of body composition. The methods were developed with clinical practice in mind and all proposed methods have been used in further studies for assessment of body composition.

En la actualidad, los avances en los sistemas de captación y procesado de imágenes médicas, permiten dotar a los clínicos de herramientas muy potentes que les ayudan en el desarrollo de su actividad clínica e investigadora. Pero no todas las herramientas comerciales permiten obtener la información necesaria para llevar a cabo ciertas investigaciones. Por esta razón surge la necesidad de realizar un procesado avanzado de ciertas imágenes médicas. Este proyecto trata de desarrollar una serie de herramientas para el análisis computacional basado en imagen cardiaca orientado fundamentalmente al escenario clínico de la insuficiencia cardiaca y la Terapia de Resincronización Cardiaca (TRC). La hipótesis global del proyecto es que el análisis más preciso de la electromecánica a partir del postprocesado de técnicas de imagen cardiaca puede mejorar los resultados clínicos de la TRC.

In this dissertation, we focus on two related parts of a 3D face recognition system with wireless transportation. In the first part, the core components of the system, namely, the feature extraction and classification component, are introduced. In the feature extraction component, range images are taken as inputs and processed in order to extract features. The classification component uses the extracted features as inputs and makes classification decisions based on trained classifiers. In the second part, we consider the wireless transportation problem of range images, which are captured by scattered sensor nodes from target objects and are forwarded to the core components (i.e., feature extraction and classification components) of the face recognition system. Contrary to the conventional definition of being a transducer, a sensor node can be a person, a vehicle, etc. The wireless transportation component not only brings flexibility to the system but also makes the “proactive” face recognition possible. For the feature extraction component, we first introduce the 3D Morphable Model. Then a 3D feature extraction algorithm based on the 3D Morphable Model is presented. The algorithm is insensitive to facial expression. Experimental results show that it can accurately extract features. Following that, we discuss the generic face warping algorithm that can quickly extract features with high accuracy. The proposed algorithm is robust to holes, facial expressions and hair. Furthermore, our experimental results show that the generated features can highly differentiate facial images. For the classification component, a classifier based on Mahalanobis distance is introduced. Based on the classifier, recognition performances of the extracted features are given. The classification results demonstrate the advantage of the features from the generic face warping algorithm. For the wireless transportation of the captured images, we consider the location-based wireless sensor networks (WSN). In order to achieve efficient routing perfor¬mance, a set of distributed stateless routing protocols (PAGER) are proposed for wireless sensor networks. The loop-free and delivery-guaranty properties of the static version (PAGER-S) are proved. Then the performance of PAGER protocols are compared with other well-known routing schemes using network simulator 2 (NS2). Simulation results demonstrate the advantages of PAGER.

Τα σύνθετα υλικά χρησιμοποιούνται σε μία ευρεία γκάμα της παγκόσμιας βιομηχανίας, στην αεροναυπηγική, στις μεταφορές, στα αθλητικά είδη και πιο πρόσφατα στις υποδομές. Αναλόγως των εφαρμογών αξιοποιούνται και αντίστοιχης ποιότητας σύνθετα υλικά διότι το κύριο μέλημα είναι η μείωση του συνολικού κόστους της κατασκευής. Η μείωση του κόστους, προφανώς, δεν έρχεται από τα ίδια τα υλικά, αφού είναι πιο ακριβά από τα αντίστοιχα που χρησιμοποιούνται τώρα, αλλά από την αξιοποίηση τους σε ένα μακροχρόνιο σχέδιο. Για τους λόγους αυτούς γίνεται όλο ένα και μεγαλύτερη χρήση των σύνθετων υλικών. Παρόλα τα πλεονεκτήματα που παρουσιάζουν τα σύνθετα υλικά εμφανίζουν πολύ σοβαρά προβλήματα. Ένα κύριο πρόβλημα το οποίο αφορά την χρήση τους είναι ο προσδιορισμός της αντοχής τους όταν υποβάλλονται σε διάφορες καταπονήσεις. Παρατηρείται λοιπόν η μεγάλη ανάγκη που υπάρχει για την μελέτη των σύνθετων υλικών και για το λόγο αυτό υπάρχουν εξειδικευμένα εργαστήρια στα οποία μελετώνται οι μηχανικές ιδιότητες των σύνθετων υλικών. Όμως, πέραν των ενδογενών μηχανικών ιδιοτήτων που εμφανίζουν τα σύνθετα υλικά υπάρχουν και άλλοι παράγοντες που χρίζουν μελέτης όπως για παράδειγμα η επίδραση της θερμοκρασίας, της υγρασίας κ.α. Τα σύνθετα υλικά που αξιοποιούνται είναι με ενίσχυση συνεχών ινών και με μήτρα από πολυμερές σε ψευδοϊσότροπη "quasi-isotropic" διαστρωμάτωση. Τα εκάστοτε δοκίμια υφίστανται εναλλασσόμενη φόρτιση μέχρις ότου εμφανίσουν αστοχία. Για την διαδικασία της κόπωσης ορίζονται πειραματικά οι τιμές της συχνότητας ταλάντωσης και της μέγιστης εφαρμοζόμενης τάσης. Όλα τα δοκίμια, κατά την διάρκεια της κόπωσης, υφίστανται μόνο τάσεις εφελκυσμού. Η μελέτη της θερμοκρασίας πραγματοποιήθηκε μέσω παθητικής θερμογραφίας. Η παθητική θερμογραφία εξετάζει τα υλικά και τις δομές που σε φυσιολογικές συνθήκες παρουσιάζουν διαφορετική (συχνά υψηλότερη) θερμοκρασία από το περιβάλλον. Η παθητική θερμογραφία είναι ευρέως χρησιμοποιούμενη σε πλήθος εφαρμογών της βιομηχανίας, της προληπτικής συντήρησης, ιατρικής, δασικής πυρανίχνευσης, γεωργίας, βιολογίας, ανίχνευσης αερίων. Σε όλες αυτές τις εφαρμογές, εντοπίζονται ασυνήθιστες θερμοκρασίες σε σχέση με το περιβάλλον και υποδεικνύουν σημεία που πιθανόν χρήζουν προσοχής. Στην παρούσα εργασία έγινε χρήση μίας πειραματικής διάταξης που είναι σε θέση να προκαλεί καταπόνηση στα σύνθετα υλικά και με τη βοήθεια υπέρυθρης κάμερας κατέστη δυνατή η απόκτηση θερμικών εικόνων των σύνθετων υλικών κατά την διάρκεια της καταπόνησης τους. Οι εικόνες αυτές αναπαριστούν την χωρική κατανομή της θερμοκρασίας του δείγματος και όχι μόνο της επιφάνειας. Οι θερμικές εικόνες υπέστησαν ειδική επεξεργασία για την εξαγωγή χαρακτηριστικών με σκοπό να οδηγήσουν σε συμπεράσματα ικανά να ερμηνεύουν την συμπεριφορά της θερμοκρασίας. Τελικά, μέσω της μελέτης της συμπεριφοράς της θερμοκρασίας των σύνθετων υλικών υφιστάμενα κόπωση έγινε προσπάθεια για την πρόβλεψη και ανίχνευση αστοχιών σε αυτά τα υλικά. / Composite materials are used in a wide range of global industry, such as aeronautics, transportation, and more recently in infrastructure. Composites with different quality are utilized because the main concern is to reduce the total cost of construction. Cost reduction is obviously not coming from the raw materials, as they are more expensive than those used now, but by exploiting them in a long-standing plan. The technological evolution and market requirements have led to an increasing demand of composite materials. Despite the advantages of composite materials they exhibit very serious problems. A fundamental problem concerning the engineering uses of composite materials is the determination of their resistance to combined states of cyclic stress (fatigue loading). Therefore, composite materials are needed to be studied and for this reason there are specialized laboratories studying their mechanical properties. Beyond the intrinsic mechanical properties of composites materials there are also other factors that need to be studied, for example the effect of temperature, humidity etc. In this project the behavior of the temperature of composites in fatigue loading is studied. The composite materials that are utilized are fiber reinforced plastics (FPR) in "quasi-isotropic" lay-up. Each sample is going under tensile constant stress cycling until experiencing failure. For the process of fatigue loading, experimental values of the oscillation frequency and the maximum applied stress are defined. The study of the temperature of composite materials was held by passive thermography. Passive thermography tests materials and structures which are naturally at different (often higher) temperature than ambient temperature. Important applications of passive thermography are in production, predictive maintenance, medicine, fire forest detection, building thermal efficiency survey programs, road traffic monitoring, agriculture and biology, medicine, detection of gas (by mean of absorbing tracer gas) and in nondestructive testing (NDT). In all these applications, abnormal temperature profiles indicate a potential problem that needs to be taken care of. In this project an experimental setup was developed which is capable of a) causing specific fatigue loading to the samples and b) obtaining thermal images of them during their fatigue loading. These images represent the spatial distribution of the sample temperature, not just the temperature of their surface. Also, these thermal images were specially processed for feature extraction in order to lead to conclusions capable of interpreting the behavior of temperature. Eventually, an effort was made to anticipate and discover failures in composites through the study of the behavior of the temperature that they develop.

Η ανάπτυξη ενός ενσωματωμένου στο όχημα συστήματος υποβοήθησης του οδηγού για αποφυγή συγκρούσεων με άλλα οχήματα, βρίσκεται τελευταία στο επίκεντρο του ενδιαφέροντος. Στα συστήματα αυτά η αξιοπιστία αποτελεί ένα πολύ σημαντικό παράγοντα. Στην παρούσα εργασία αναπτύσσεται ένα σύστημα αναγνώρισης προπορευόμενου οχήματος βασισμένο σε εικόνες οι οποίες λαμβάνονται από βιντεοκάμερα που έχει ενσωματωθεί στο όχημα. Η μεθοδολογία την οποία επιλέξαμε να εργαστούμε περιλαμβάνει τον εντοπισμό των κόκκινων εικονοστοιχείων στην εικόνα και τη δημιουργία της αντίστοιχης δυαδικής εικόνας. Στη συνέχεια, με μορφολογική επεξεργασία της δυαδικής εικόνας εντοπίζουμε τις περιοχές που αντιστοιχούν στα πιθανά φανάρια του οχήματος. Με βάση τα σημεία των πιθανών φαναριών καθορίζουμε την περιοχή στην οποία περικλείεται το όχημα. Για την επιβεβαίωση της ύπαρξης οχήματος στην περιοχή αυτή, εκτελούμε έναν έλεγχο συμμετρίας βασιζόμενοι στην ομοιότητα των υποεικόνων και συνεχίζουμε με τον προσεγγιστικό υπολογισμό της απόστασής του. Τέλος, παρουσιάζουμε τα αποτελέσματα της μεθόδου, τα συμπεράσματα που προέκυψαν και προτείνουμε κατευθύνσεις για μελλοντικές βελτιώσεις. / Developing on-board automotive driver assistance systems aiming to alert drivers about possible collision with other vehicles has attracted a lot of attention lately. In these systems, robust and reliable vehicle detection is a critical step. In this work a vehicle detection system is developed based on video frames grabbed by a camera mounted on the vehicle. Vehicle detection is mainly based on the detection of its red rear-lights. First we detect all red pixels of the frame and create the corresponding binary image (mask). Then we detect the areas that possibly constitute vehicle’s rear-lights by performing morphological binary image processing. Based on that, we determine the boundary of the vehicle. To verify the presence of the vehicle in this area, we perform a symmetry test based on sub-image similarity. Finally, we present some experimental results and give directions for future improvements.

Στα πλαίσια της διατριβής μελετήθηκε αρχικά το πρόβλημα της εύρεσης της θέσης του υφάσματος επάνω στην τράπεζα εργασίας με μεθόδους ψηφιακής ανάλυσης σημάτων βίντεο, παρουσία φαινομένων παραμόρφωσης και μερικής επικάλυψης του υφάσματος. Οι νέες μέθοδοι εντοπισμού που αναπτύχθηκαν αξιολογήθηκαν πειραματικά παρουσιάζοντας ικανοποιητική ακρίβεια εντοπισμού και ανοχή του συστήματος σε μερικές επικαλύψεις και παραμορφώσεις. Μετά την ολοκλήρωση της ραφής του υφάσματος πραγματοποιείται αυτόματος εντοπισμός της θέσης της ραφής από ψηφιακές φωτογραφίες. Αναπτύχθηκαν τρείς πρωτότυπες μέθοδοι εντοπισμού της θέσης της ραφής οι οποίες διαφοροποιούνται στο στάδιο της προεπεξεργασίας. Η πειραματική αξιολόγηση γίνεται σε βάση δεδομένων που περιέχει 118 εικόνες έτοιμων ενδυμάτων. Προτού πραγματοποιηθεί ποιοτικός έλεγχος ραφής, οι εικόνες κανονικοποιούνται ως προς τη θέση και τον προσανατολισμό της ραφής χρησιμοποιώντας τις παραπάνω μεθόδους αυτόματου εντοπισμού της θέσης της ραφής. Αναπτύχθηκαν και αξιολογήθηκαν τρείς διαφορετικές μέθοδοι αυτόματης αναγνώρισης της ποιότητας σε δείγματα ραφής οι οποίες εξάγουν τρία διαφορετικά σύνολα χαρακτηριστικών. Η πρώτη μέθοδος βασίζεται σε φασματικά χαρακτηριστικά, η δεύτερη στην επιβολή αυτό-σκίασης, ενώ η τρίτη βασίζεται στην εκτίμηση της ανομοιομορφίας της επιφάνειας των δειγμάτων ραφής. Η πειραματική αξιολόγηση γίνεται σε βάση δεδομένων δειγμάτων ραφής που περιλαμβάνει 325 ραφές. Η εκτίμηση της ποιότητας ραφής πραγματοποιείται με ταξινόμηση σε πέντε διατεταγμένους βαθμούς ποιότητας. Σε αυτήν την κατεύθυνση, προτείνονται και συγκρίνονται τέσσερις μέθοδοι αναγνώρισης προτύπων διατεταγμένων κατηγοριών. Η πρώτη μέθοδος χρησιμοποιεί για την ταξινόμηση μοντέλο σύμμετρων αναλογιών πιθανότητας. Η δεύτερη μέθοδος κάνει αναγνώριση με χρήση γραμμικού μοντέλου. Οι άλλες δύο μέθοδοι είναι πρωτότυπες και επίσης χρησιμοποιούν γραμμικό μοντέλο για την ταξινόμηση. Η διαφοροποίησή τους από τη δεύτερη μέθοδο είναι ότι η επιλογή των αριθμητικών τιμών των διατεταγμένων κατηγορίων δεν γίνεται αυθαίρετα., αλλά προκύπτει ως λύση προβλημάτων ελαχιστοποίησης.. Η πειραματική αξιολόγηση και σύγκριση των μεθόδων στο πρόβλημα του ποιοτικού ελέγχου ραφών οδηγεί στην επιλογή του μοντέλου σύμμετρων αναλογιών πιθανότητας σε περίπτωση που υπάρχει ικανός αριθμός παραδειγμάτων εκπαίδευσης, ενώ σε αντίθετη περίπτωση μπορεί να προτιμηθεί το γραμμικό μοντέλο αφού προηγηθεί βελτιστοποίηση με χρήση κάποιας εκ των δύο προτεινόμενων μεθόδων επιλογής αριθμητικών τιμών. / One of the problems studied in the present dissertation is that of the detection of the fabrics’ position on the working area. The proposed detection methods are based on image processing and analysis techniques and take into consideration both partial occlusion and fabric deformation. The methods have been experimentally evaluated and the results indicate sufficient detection accuracy and robustness regarding partial occlusion and fabric deformation. After sewing the fabric, the position and orientation of the seam is automatically detected. Three novel seam detection methods have been developed using different pre-processing techniques. The experimental evaluation of the three detection methods is made on a database containing 118 images of ready sewn garments. Before performing seam quality control the seam images are normalized with respect to the seam position and orientation, using the aforementioned seam detection methods. Feature selection has been studied next, extracting three different sets of features and assessing seam quality using three different methods. The first method uses spectral features; the second method is based on the detection of self-shadows onto the seam specimens, while the third method is based on the estimation of the surface roughness of the specimens. The experimental evaluation of the proposed methods is made on a database containing 325 images of seam specimens. Seam quality control is performed by classifying the seam specimens into five ordinal grades of quality. In this direction, four classification methods are proposed and evaluated, taking into account the ordered arrangement of the classes. The first method uses the proportional odds model; while the second method uses a linear model. The other two methods are novel and also employ a linear model. The difference between these two methods and the second method is that the numerical values they are assigning to the ordered categories are not arbitrary like in the case of the second method. The experimental evaluation of these four methods indicates that in case of a large number of training data, the first method which is based on the proportional odds model is more efficient, while in case of an insufficient number of training data the linear model optimized by one of the two novel methods should be selected.

Primarily due to increasing security demands and potential commercial and law enforcement applications, automatic face recognition has been a subject of extensive study in the past several decades, and remains an active field of research as of today. As a result, numerous techniques and algorithms for face recognition have been developed, many of them proving effective in one way or another. Nevertheless, it has been realized that constructing good solutions for automatic face recognition remains to be a challenge. The last two decades have witnessed significant progress in the development of new methods for automatic face recognition, some being effective and robust against pose, illumination and facial expression variations, while others being able to deal with large-scale data sets. On all accounts, the development of state-of-the-art face recognition systems has been recognized as one of the most successful applications of image analysis and understanding. Among others, the principal component analysis (PCA) developed in the early 1990s has been a popular unsupervised statistical method for data analysis, compression and visualization, and its application to face recognition problems has proven particularly successful. The importance of PCA consists in providing an efficient data compression with reduced information loss, and efficient implementation using singular value decomposition (SVD) of the data matrix. Since its original proposal, many variations of the standard PCA algorithm have emerged. This thesis is about enhancement and extensions of the standard PCA for face recognition. Our contributions are twofold. First, we develop a set of effective pre-processing techniques that can be employed prior to PCA in order to obtain improved recognition rate. Among these, a technique known as perfect histogram matching (PHM) is shown to perform very well. Other pre-processing methods we present in this thesis include an extended sparse PCA algorithm for dimensionality reduction, a wavelet-transform and total variation minimization technique for dealing with noisy test images, and an occlusion-resolving algorithm. Second, we propose an extended two-dimensional PCA method for face recognition. This method, especially when combined with a PHM pre-processing module, is found to provide superior performance in terms of both recognition rate and computational complexity.