Ce mémoire aborde le problème de la détection de visages à partir d'une image compressée. Il touche également à un problème connexe qui est la qualité des standards de compression et l'estimation de celle-ci. Ce mémoire est organisé sous la forme d'une introduction générale sur la détection de visages et de deux articles soumis à des conférences internationales. Le premier article propose une amélioration de la méthode classique pour comparer la qualité de deux standards. Le deuxième propose une méthode de décompression spécialisée pour faire fonctionner le détecteur de visages de Viola-Jones dans le domaine compressé.

The master thesis presents an overview of face detection task in color, static images. Face detection term is posed in the context of various branches. Main concepts of face detection and also their relationships are described. Individual approaches are divided into groups and then define in turn. In the thesis is in detail described algorithm AdaBoost, which is selected on the basis of its properties. Especially speed of computation and good detection results are key features. In the scope of this work Viola-Jones detector was implemented. This detector was trained with face pictures from public accessible database. Combination of Viola-Jones detector with simple color detector is described. In the thesis is also presented experiment approach to facial features detection.

Ingeniero Civil Eléctrico / La revisión sistemática de la prensa es una importante herramienta para el análisis de presencia de marcas y desarrollo de estrategias publicitarias, el monitoreo de estos medios tradicionalmente es realizado por operadores que manualmente extraen la información. Este trabajo tiene como objetivo presentar un novedoso sistema de detección automática de publicidad sobre imágenes procedentes de diarios y revistas, dentro del marco del proyecto IntelliMEDIA, llevado a cabo por el startup chileno CPDLabs. La metodología para detectar anuncios publicitarios se basa en un modelamiento específico de la estructura de la prensa escrita, permitiendo llegar a obtener una estimación del costo de los anuncios detectados. Para ello se separa el problema en 4 bloques: Preprocesamiento que separa el texto de las imágenes dentro de la página. Detección de logos que busca logos dentro de la página. Detección de publicidad que identifica el anuncio publicitario al que pertenece el logo y finalmente una etapa de tarificación que entrega una estimación del costo asociado al espacio publicitario. El problema es abordado principalmente mediante tres estrategias: Primero la representación de la imagen en descriptores locales, que permite calzar características similares entre imágenes. Segundo, la estrategia de detección de objetos Viola-Jones, algoritmo de machine learning que genera un clasificador en base a un conjunto de imágenes de entrenamiento. La última estrategia es comparar histogramas de color permitiendo integrar información de color a la clasificación. Para medir el desempeño de dichas estrategias se desarrolla un marco de evaluación, que consiste en una base de validación de 20.000 páginas de diario con 27 logos marcados, para así medir el desempeño de las distintas estrategias y configuraciones de parámetros, encontrar una solución eficaz para el problema y analizar las fortalezas y debilidades de los distintos métodos. Los resultados demuestran que la solución es viable y es posible detectar logos mediante descriptores locales y Viola-Jones, logrando desempeños mayores al 90%. Por lo tanto IntelliMediA puede llegar a ser una manera eficaz y eficiente de extraer información publicitaria automáticamente de prensa escrita.

Rip current images are useful for assisting in climate studies but time consuming to manually annotate by hand over thousands of images. Object detection is a possible solution for automatic annotation because of its success and popularity in identifying regions of interest in images, such as human faces. Similarly to faces, rip currents have distinct features that set them apart from other areas of an image, such as more generic patterns of the surf zone. There are many distinct methods of object detection applied in face detection research. In this thesis, the best fit for a rip current object detector is found by comparing these methods. In addition, the methods are improved with Haar features exclusively created for rip current images. The compared methods include max distance from the average, support vector machines, convolutional neural networks, the Viola-Jones object detector, and a meta-learner. The presented results are compared for accuracy, false positive rate, and detection rate. Viola-Jones has the top base-line performance by achieving a detection rate of 0.88 and identifying only 15 false positives in the test image set of 53 rip currents. The described meta-learner integrates the presented Haar features, which are developed in accordance with the original Viola-Jones algorithm. Ada-Boost, a feature ranking algorithm, shows that the newly presented Haar features extract more meaningful data from rip current images than some of the current features. The meta-classifier improves upon the stand-alone Viola-Jones when applying these features by reducing its false positives by 47% while retaining a similar computational cost and detection rate.

Thesis (MSc)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: In South Africa, black-back jackal (BBJ) predation of sheep causes heavy losses to sheep farmers. Different control measures such as shooting, gin-traps and poisoning have been used to control the jackal population; however, these techniques also kill many harmless animals, as they fail to differentiate between BBJ and harmless animals. In this project, a system is implemented to detect black-backed jackal faces in images. The system was implemented using the Viola-Jones object detection algorithm. This algorithm was originally developed to detect human faces, but can also be used to detect a variety of other objects. The three important key features of the Viola-Jones algorithm are the representation of an image as a so-called ”integral image”, the use of the Adaboost boosting algorithm for feature selection, and the use of a cascade of classifiers to reduce false alarms. In this project, Python code has been developed to extract the Haar-features from BBJ images by acting as a classifier to distinguish between a BBJ and the background. Furthermore, the feature selection is done using the Asymboost instead of the Adaboost algorithm so as to achieve a high detection rate and low false positive rate. A cascade of strong classifiers is trained using a cascade learning algorithm. The inclusion of a special fifth feature Haar feature, adapted to the relative spacing of the jackal’s eyes, improves accuracy further. The final system detects 78% of the jackal faces, while only 0.006% of other image frames are wrongly identified as faces. / AFRIKAANSE OPSOMMING: Swartrugjakkalse veroorsaak swaar vee-verliese in Suid Afrika. Teenmaatreels soos jag, slagysters en vergiftiging word algemeen gebruik, maar is nie selektief genoeg nie en dood dus ook vele nie-teiken spesies. In hierdie projek is ’n stelsel ontwikkel om swartrugjakkals gesigte te vind op statiese beelde. Die Viola-Jones deteksie algoritme, aanvanklik ontwikkel vir die deteksie van mens-gesigte, is hiervoor gebruik. Drie sleutel-aspekte van hierdie algoritme is die voorstelling van ’n beeld deur middel van ’n sogenaamde integraalbeeld, die gebruik van die ”Adaboost” algoritme om gepaste kenmerke te selekteer, en die gebruik van ’n kaskade van klassifiseerders om vals-alarm tempos te verlaag. In hierdie projek is Python kode ontwikkel om die nuttigste ”Haar”-kenmerke vir die deteksie van dié jakkalse te onttrek. Eksperimente is gedoen om die nuttigheid van die ”Asymboost” algoritme met die van die ”Adaboost” algoritme te kontrasteer. ’n Kaskade van klassifiseerders is vir beide van hierdie tegnieke afgerig en vergelyk. Die resultate toon dat die kenmerke wat die ”Asymboost” algoritme oplewer, tot laer vals-alarm tempos lei. Die byvoeging van ’n spesiale vyfde tipe Haar-kenmerk, wat aangepas is by die relatiewe spasieëring van die jakkals se oë, verhoog die akkuraatheid verder. Die uiteindelike stelsel vind 78% van die gesigte terwyl slegs 0.006% ander beeld-raampies verkeerdelik as gesigte geklassifiseer word.

The area of computer vision has been making big improvements in the latest decades, equally so has the area of electronics and small computers improved. These areas together have made it more available to build small, standalone systems for object detection in live video. This project's main objective is to examine whether a small device, e.g. Raspberry Pi 3, can manage an implementation of an object detection algorithm, called Viola-Jones, to count the occupancy of sitting people in a room with a camera. This study is done by creating an application with the library OpenCV, together with the language C+ +, and then test if the application can run on the small device. Whether or not the application will detect people depends on the models used, therefore three are tested: Haar Face, Haar Upper body and Haar Upper body MCS. The library's object detection function takes some parameters that works like settings for the detection algorithm. With that, the parameters needs to be tailored for each model and use case, for an optimal performance. A function was created to find the accuracy of different parameters by brute-force. The test showed that the Haar Face model was the most accurate. All the models, with their most optimal parameters, are then speed-tested with a FPS test on the raspberry pi. The result shows whether or not the raspberry pi can manage the application with the models. All models could be run and the Haar face model was fastest. As the system uses cameras, some ethical aspects are discussed about what people might think of top-corner cameras.

The aim of this diploma thesis is to explore methods of face detection and recognition in the picture. The method for face detection and the method for face recognition will be chosen according to literature survey. Both methods will be implemented using the OpenCV library and a program language C/C++. The result of this project is creation of graphic interface which use programmed function for face detection and recognition from a picture and also a camcorder.

This diploma thesis deals with face recognition in digital pictures. The first part describes biometry and, shortly, characterizes biometrical methods which are the most oftenly used. In the second part is described the approach of face recognition in a picture. Specifically, it is described the method for face detection - Viola-Jones and method for face recognition - PCA, which will be implemented in Matlab. The last part, which is practical, describes the scheme for video-sequence recording, implementation of the PCA method in Matlab and discussion of the achieved results.

No / In this paper we present a new method for accurate real-time facial feature detection. Our method is based on local feature detection and enhancement. Previous work in this area, such as that of Viola and Jones, require looking at the face as a whole. Consequently, such approaches have increased chances of reporting negative hits. Furthermore, such algorithms require greater processing power and hence they are especially not attractive for real-time applications. Through our recent work, we have devised a method to identify the face from real-time images and divide it into regions of interest (ROI). Firstly, based on a face detection algorithm, we identify the face and divide it into four main regions. Then, we undertake a local search within those ROI, looking for specific facial features. This enables us to locate the desired facial features more efficiently and accurately. We have tested our approach using the Cohn-Kanade’s Extended Facial Expression (CK+) database. The results show that applying the ROI has a relatively low false positive rate as well as provides a marked gain in the overall computational efficiency. In particular, we show that our method has a 4-fold increase in accuracy when compared to existing algorithms for facial feature detection.

No / The purpose of this chapter is to discuss and present a computational framework for detecting and analysing facial expressions efficiently. The approach here is to identify the face and estimate regions of facial features of interest using the optical flow algorithm. Once the regions and their dynamics are computed a rule based system can be utilised for classification. Using this framework, we show how it is possible to accurately identify and classify facial expressions to match with FACS coding and to infer the underlying basic emotions in real time.