Human Face Detection Improvement Using Incremental Learning Based on Low Variance Directions

Fast face detection using a unified architecture for unconstrained and infrared face images

Fatigue driving detection helps reduce the occurrence of traffic accidents. Currently, the fatigue detection methods still have following challenges: 1) In the visual fatigue detection system, face detection is the foundation of fatigue detection and it is always susceptible to changes of illumination and facial expression, which lowers the accuracy of face detection. 2) It is so difficult to distinguish some behaviors of a driver such as singing and yawning in a single frame that making false fatigue detection in the visual fatigue detection system. To overcome the problems mentioned above, a fatigue detection method based on sequential mouth feature with long short-term memory network is proposed. The proposed method mainly consists of two parts: locating of facial landmark based on face and mouth detection, and yawning detection based on residual network (Resnet) and long short-term memory (LSTM). The face detection method is based on Histogram of Oriented Gradient (HOG) feature and Support Vector Machine (SVM) classifier, which is robust to the changes of illumination and slightly facial expression. The mouth detection is based on cascaded regression tree, which extracts the mouth region from the detected face. The Resnet and LSTM based yawning detection extracts spatial-temporal features from the mouth regions between frames and distinguishes the behaviors about yawning to identify fatigue of a driver with sequential mouth feature. The proposed method achieves accuracy of 94.9% on the YawDD dataset, which outperforms the compared methods. Experimental results demonstrate that the proposed method satisfies the performance of fatigue detection.

Thermal imagery is a substitute of visible imagery for face detection due to its property of illumination invariance with the variation of facial appearances. This paper presents an effective method for human face detection in thermal imaging. The concept of histogram plot has been used in the feature extraction process and later in face detection. Techniques like thresholding, object boundary analysis, morphological operation etc. have been performed on the images to ease the process of detection. In order to enhance the performance of the algorithm and to reduce the computation time, parallelism has been achieved using Message Passing Interface (MPI) model. Overall, the proposed algorithm showed a higher level of accuracy and less complexity time of 0.11 seconds in the parallel environment as compared to 0.20 seconds in a serial environment.

Face and eye detection is one of the most challenging problems in computer vision area. The goal of this paper is to present a study over the existing literature on face and eye detection and gaze estimation. With the uptrend of systems based on face and eye detection in many different areas of life in recent years, this subject has gained much more attention by academic and industrial area. Many different studies have been performed about face and eye detection. Besides having many challenging problems like, having different lighting conditions, having glasses, facial hair or mustache on face, different orientation pose or occlusion of face, face and eye detection methods performed great progress. In this paper we first categorize face detection models and examine the basic algorithms for face detection. Then we present methods for eye detection and gaze estimation.

In this paper, a method for detecting the direction of a human face is developed; regardless of its age or sex. The method involves creating a set of five face patterns representing the front, up, down, left, and right directions of a face. The face patterns are produced by applying Canny’s edge detection algorithm on some face files. The direction of the input face is found by first applying the above algorithm on the input file and comparing it with the five face patterns. The face pattern that gives minimum difference will represent the direction of the input face. Excellent results were reported when applied on images with relatively clear background and the head were centered at the image area.

This paper presents a novel face detection method by applying distribution-based distance (DBD) measure and support vector machine (SVM). The novelty of our DBD-SVM method comes from the integration of discriminating feature analysis, face class modeling, and support vector machine for face detection. First, the discriminating feature vector is defined by combining the input image, its 1-D Haar wavelet representation, and its amplitude projections. Then the DBD-SVM method statistically models the face class by applying the discriminating feature vectors and defines the distribution-based distance measure. Finally, based on DBD and SVM, three classification rules are applied to separate faces and nonfaces. Experiments using images from the MIT-CMU test sets show the feasibility of our new face detection method. In particular, when using 92 images (containing 282 faces) from the MIT-CMU test sets, our DBD-SVM method achieves 98.2% correct face detection accuracy with 2 false detections, a performance comparable to the state-of-the-art face detection methods, such as the Schneiderman-Kanade's method.

This paper proposes a method of evolutionary robot vision based on a local genetic algorithm based on clustering (LGAC) and multi-stage fuzzy evaluation. In order to improve the communication capability of human-friendly partner robots, the human face should be extracted as correctly as possible. First, we discuss the concept of evolutionary robot vision in dynamic environments. Next, we propose growing neural gas for preprocessing as a bottom–up processing, and local genetic algorithm based on clustering for template matching in human face detection as a top–down processing. In order to improve the performance of the human face detection, we use multi-stage fuzzy evaluation for evaluating the degree of human face. Finally, we show several experimental results and discuss the effectiveness of the proposed method.

Face detection methods are in general applied frame by frame based on face characteristics. Human faces are part of a living body in which there is a constant blood flow. It has been recently demonstrated that heart rate could be estimated with simple consumer level digital cameras. In this paper, we propose a new method to detect faces in a video stream based on the detection of heart rate. Images are sliced into small regions of interest. For each region, heart rate is estimated and an index of presence is computed. A face detection map can be created by using the index of presence and the heart rate information. This system could be used to increase the robustness of face detection systems based on face characteristics only, emotion recognition or biometric systems.

The face verification systems gained significant attention in the last few years due to the increased security concern in public and private places. Face detection is the most important and initial stage in the automatic face verification system. It helps to determine the existence of faces in an image and return the position and location of the face. The face verification system's accuracy depends on face detection. The human faces are not always frontal and have many variations, therefore, face detection is challenging in unconstrained scenarios. One main challenge of face detection is occlusion. The proposed work is an attempt to illustrate the cognitive informatics approach using machine learning and present an occluded face detection method. The proposed method uses Adaboost[1] machine learning approach. The Viola-Jones[2] algorithm along with free rectangular features[3] has been adopted in the proposed approach in order to detect faces. the machine learning methods require two operation namely training and testing. Two cascade classifiers are used in which one is trained on holistic faces and the second is trained on half occluded faces; both of the classifiers are used in parallel to work in unconfined scene. Additionally, for improvement the correctness and adeptness of the system, the skin color models are applied which are used for removing of the false positive detection. The experiment has been performed on FDDB[4] dataset. The results shows that the proposed method achieve desirable results in the detection of half occluded faces.

Automatic face detection in video sequences using local normalization and optimal adaptive correlation techniques

Face detection is a common computer technology being used in human identification applications. It can also refer to the process of locating human faces in a visual scene. Face detection is a branched field of object detection where all objects in an image are detected including several classes like cars, trees, humans… etc. Also face detection problems branch into a lot of cases, some focus on frontal faces, others focus on side pose and so on. In this paper, a new face detection method based on Bilinear Interpolation image zooming method and image enhancement by Adaptive Histogram Equalization (AHE) method is proposed. The new method gives an encouraging results for crowded human images. By comparing the proposed method with the Viola-Jones algorithm, face detector using the cascade object detector, which supported in MATLAB, the new method gives excellent results in detecting human faces with different resolutions, poses and sizes. It succeeds in detecting most of the human faces in the tested images regardless of image sizes. The new method is tested on several images in Pratheepan dataset with crowded humans. Also, I tested the new method on many images collected from the Internet, whose can be classified as crowded human images. Experimental results show that the proposed Ad_L_Hist method is more efficient in detecting human faces in crowded human images.

During the past few years face detection and tracking from video has received maximum importance because of commercial and enforcement application of varied an extreme range. It is also most challenging task in video, where the variation of illuminations, noise, locations of human face and pose can differ from one frame to another. For face detection and tracking from video database is being presented a unique technique in this paper. In this study, primary goal is to recognize location of faces from video. Moreover, finding face motion leads to be a part of face recognition system. Firstly, face edges are detected using Robert edge detector followed by a set of arithmetic operations between an initial frame and the nearest ones. Thereafter, non-desired edges and noise are removed by Gaussian filtering technique. A logical operation is then performed between the previous two output frames and noiseless face contour frame for detecting edges corresponding to face video. Finally, four corner points i.e. top-left, top-right, bottom-left, bottom-right are computed to draw rectangle around the face and detect face contour of each frame. To track human face from video, scalar and vector distance between four corner points of two consecutive frames are calculated. Displacement of corner points means position and location of face changes in the next frame. On Honda/UCSD video database the proposed method has been tested and it has been found through experimental results that it can detect and track from video efficiently human face.

Nowadays, face detection is common. Its used in many areas. With the help of face detection benchmark data-sets, many signs of progress have been made. Face detection methods used nowadays is not matching the real-world requirements. With new advancing technologies and services, we need to upgrade our existing systems. By using a data-set called as WIDER FACE which is very large in size than already existing data-sets, we can improve the performance. WIDER FACE data-set has many faces in it which may be challenging as it includes faces under different conditions. Moreover, we can see that in face detection task, WIDER FACE data-set is best for training and testing the accuracy of the model. But existing face detection algorithms and models are not up-to the mark. They have major limitations. So we created a WIDER FACE detection system which will help us overcome all those issues.

There are many theft cases which happen because there is no tool used as a guard. Based on the situation mentioned, the house owner tends to use a surveillance camera to guard their properties, especially an important area such as house rooms. However, in general, the traditional surveillance camera is a passive device which has no notification feature to inform the house owner. Indeed, it also has no ability to recognize a human existence. So that, the object detection related to the human and also the notification corresponds a criminal action which is going to be acted by the suspected human is really needed. According to the object detection as the human being, there is an image processing needed to treat the picture which is captured by the surveillance camera. There are many ways to improve the object detection performance of the surveillance camera. One of them is by making a captured image sample which is caught by the surveillance camera. Firstly, the human is a moving object. So that, it is necessary to select the method which is related to the moving object detection to make a sampled image. One of the methods which can be applied is the Background Subtraction. When the moving target is known and is sampled, then the next step is by separating the human by classifying it from the object. In this study, human detection is generated by using Face Detection for the first mechanism and for the second mechanism is Head and Shoulders Detection. Both mechanisms are formed in master/slave. Thus, if the master function which is the Face Detection cannot recognize the object as a human, then the Head and Shoulders detection is going to be shown. The inability of master function for detection happens because of less illumination, or an obstacle such as the human face appears in the opposite direction from the lens of the camera. To minimize the detection failure, the night vision feature is also presented in this study for the surveillance camera prototyping. The human object detection by using the Head and Shoulders method as slave feature is still established for minimizing the failure probability. The proposed system works properly, especially when the system uses the combination of Face Detection and also Head and Shoulders method. The usage of night vision feature also improves the accuracy for detecting the object moves more than 80% in several variations of illumination.

In the current scenario face identification and recognition is an important technique in surveillance. The face is a necessary biometric in humans. Therefore face detection plays a major job in computer vision applications. Several face recognition and emotions classification approaches have been presented throughout the last few decades of research to improve the rate of face recognition for thermal pictures. However, in real-time, lighting conditions might change due to several factors, such as the different times of capture, weather, etc. Due to variations in lighting intensity, the performance of the facial expression recognition system is not good. This paper proposed a model for human thermal face detection and expression classification. Four main steps were involved in this research. Initially, the Difference of the Gaussian (DOG) filter is utilized to crop the input thermal images and then normalize the images using the median filter in pre-processing step. Then, Efficient Net is used for extracting features such as shape, location, and occurrences from thermal face images. After that, detect human faces utilized by the YOLOv4 technique to better emotions classification. Finally, classify the emotions on faces by using the DenseNet technique into seven emotions such as happy, sad, disgust, surprise, anger, fear, and neutral. The proposed method outperforms state-of-art techniques for face recognition on thermal pictures, and classifies the expressions, according to experimentations on the RGB-D-T database. The accuracy, precision, recall, and f1-score metrics will be utilized with the database to assess the efficacy of the proposed methodology. The proposed models achieve a high classification accuracy of 95.97% on the RGB-D-T database. Furthermore, the outcomes show good precision for various face recognition tasks.