Detection of Humans Using Color Information

نویسنده

  • Vı́t Novák
چکیده

In this report a method for detection of human faces from color images based on human skin color detection and subsequent segmentation and verification scheme is described. Using the large Compaq database, skin and non-skin color models are build on the basis of simple histogram approximation of the probability distributions and the Neyman-Pearson classifier used in [8] for pixel-level skin detection is designed. A connected component analysis for reducing the detection errors is proposed, resulting in a slight improvement of ROC curve in comparison with [8]. Using this method and the output mask from a skin classifier, the segmentation algorithm based on a region growing technique using the color distribution is applied. The set of face candidates obtained from the region growing algorithm is analyzed in a very simple rejecting scheme. We show respectable results from the face detection algorithm, although the face shape model is very primitive. Anotace Tato práce popisuje metodu detekce obličejů v barevných obrázcı́ch. Hlavnı́ prvky metody spočı́vajı́ v detekci kůže, segmentaci homogennı́ch oblastı́ a následné verifikaci jednotlivých oblastı́ . K detekci kůže jsou použity dva modely. Histogram barvy kůže a histogram ostatnı́ch barev (nekůže). Tyto modely sloužı́ jako aproximace rozloženı́ pravděpodobnosti barvy kůže resp. nekůže s jejichž pomocı́ je navržen Neyman-Pearsonův klasifikátor barvy kůže poprvé implementovaný v [8]. Výstup klasifikátoru je použit pro následnou segmentaci, která využı́vá techniku narůstánı́ založenou na barvě a prostorovém rozloženı́. Výsledná množina kandidátů je postupně ověřena pomocı́ primitivnı́ho prostorového modelu obličeje. Přestože zmı́něný model obličeje je velmi jednoduchý, výsledný algoritmus vykazuje slibné výsledky. I would like to express my thanks to Mr. Jones who provided us with the Compaq image database and especially to my supervisor Mr. Matas for his time and a great deal of help he gave me.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Detection of Blood Vessels in Color Fundus Images using a Local Radon Transform

Introduction: This paper addresses a method for automatic detection of blood vessels in color fundus images which utilizes two main tools: image partitioning and local Radon transform. Material and Methods: The input images are firstly divided into overlapping windows and then the Radon transform is applied to each. The maximum of the Radon transform in each window corresponds to the probable a...

متن کامل

A Saliency Detection Model via Fusing Extracted Low-level and High-level Features from an Image

Saliency regions attract more human’s attention than other regions in an image. Low- level and high-level features are utilized in saliency region detection. Low-level features contain primitive information such as color or texture while high-level features usually consider visual systems. Recently, some salient region detection methods have been proposed based on only low-level features or hig...

متن کامل

The Combinational Use Of Knowledge-Based Methods and Morphological Image Processing in Color Image Face Detection

The human facial recognition is the base for all facial processing systems. In this work a basicmethod is presented for the reduction of detection time in fixed image with different color levels.The proposed method is the simplest approach in face spatial localization, since it doesn’trequire the dynamics of images and information of the color of skin in image background. Inaddition, to do face...

متن کامل

Using P300 to Evaluate the Effect of Object Color Knowledge in Novelty Detection

A B S T R A C T Introduction: In an oddball experiment, the context in which novel stimuli are presented affects characteristics of novelty P3, i.e. as long as there is a difficult task in which the difference between standard and target stimuli is small, recurrent presentation of a highly discrepant stimulus can lead to P300 highly similar to novelty P3. Effect of stimulus properties on P300 h...

متن کامل

Face Detection with methods based on color by using Artificial Neural Network

The face Detection methodsis used in order to provide security. The mentioned methods problems are that it cannot be categorized because of the great differences and varieties in the face of individuals. In this paper, face Detection methods has been presented for overcoming upon these problems based on skin color datum. The researcher gathered a face database of 30 individuals consisting of ov...

متن کامل

A New Method for Eye Detection in Color Images

The problem of eye detection in face images is very important for a large number of applications ranging from face recognition to gaze tracking. In this paper we propose a new algorithm for eyes detection. First, the face region is extracted from the image by skin-color information. Second, horizontal projection in image is used to approximate region of the eye be obtained . At last, the eye ce...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2001