Global Haar-Like Features: A New Extension of Classic Haar Features for Efficient Face Detection in Noisy Images

Rezaei, Mahdi; Nafchi, Hossein Ziaei; Morales, Sandino

doi:10.1007/978-3-642-53842-1_26

Cited by 14 publications

(13 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For instance, the haar cascade arithmetic [12] is used to detect single or multiple faces and both eyes. To automatically find out and track the eye pupils' movement, the object detection [14], edge detection [9], and pattern recognition [8] are further employed in this study.…”

Section: Previous Workmentioning

confidence: 99%

See 1 more Smart Citation

Prototyping An Autonomous Eye-Controlled System (AECS) using Raspberry-Pi on Wheelchairs

Thota¹,

Vangali²,

Yang³

2017

IJCA

View full text Add to dashboard Cite

In order to help physically disabled persons to make their life independent, this paper proposes an autonomous eye controlled system (AECS) on wheelchairs. In this work, several OpenCV image processing algorithms are employed to track the eye motion to coordinate the wheelchair moving left, right, and straight forward. We use the Raspberry-Pi B+ board as the system center to process the images and control the motors via GPIO. Experimental results show that the ACES system can be effectively used in the prototype, and outperforms the hand gesture controlled system by 25% processing latency reduction. General TermsComputer engineering, embedded system design

show abstract

Section: Previous Workmentioning

confidence: 99%

“…In this study, the haar course calculation is utilized for the face and eye identification [12]. After recognizing the faces, it attempts to identify the eye inside the face area and draws the rectangular box over the eyes.…”

Section: Face and Eye Detectionmentioning

confidence: 99%

Prototyping An Autonomous Eye-Controlled System (AECS) using Raspberry-Pi on Wheelchairs

Thota¹,

Vangali²,

Yang³

2017

IJCA

View full text Add to dashboard Cite

show abstract

“…Haar features have been widely used in the detection of faces [11][12][13] pedestrians [14], gestures [15,16], text within images [17], vehicles [18,19], among others. These features encode the mean intensities differences between two adjacent rectangular regions in an image, showing structural similarities between objects.…”

Section: Haar Featuresmentioning

confidence: 99%

Body Gestures Recognition System to Control a Service Robot

Medina-Catzin¹,

Martín-González²,

Brito‐Loeza³

et al. 2017

IJITCS

View full text Add to dashboard Cite

Abstract-Personal service robots will be in the short future part of our world by assisting humans in their daily chores. A highly efficient way of communication with people is through basic gestures. In this work, we present an efficient body gestures' interface that gives the user practical communication to control a personal service robot. The robot can interpret two body gestures of the subject and performs actions related to those gestures. The service robot's setup consists of a Pioneer P3-DX research robot, a Kinect sensor and a portable workstation. The gesture recognition system developed is based on tracking the skeleton of the user to get the body parts relative 3D positions. In addition, the system takes depth images from the sensor and extracts their Haar features, which will train the Adaboost algorithm to classify the gesture. The system was developed using the ROS framework, showing good performance during experimental evaluation with users. Our body gesturebased interface may serve as a baseline to develop practical and natural interfaces to communicate with service robots in the near future.

show abstract

“…By extending our previous work, originally developed for face and eye classification [37], [38], we propose a vehicle classifier which is adaptive with respect to fast intensity changes and extreme lighting conditions to ensure successful vehicle detection at day or night, also under challenging lighting. Furthermore, we develop a new training phase to create a boosted cascade of weak classifiers based on recently proposed global Haar features [36], as an efficient complement for standard Haar features. Both contributions together lead to a faster and more accurate classifier, outperforming the standard classifiers.…”

Section: Adaptive Global Haar Classifiermentioning

confidence: 99%

Robust Vehicle Detection and Distance Estimation Under Challenging Lighting Conditions

Rezaei

Terauchi

Klette

2015

IEEE Trans. Intell. Transport. Syst.

Self Cite

115

View full text Add to dashboard Cite

Avoiding high computational costs and calibration issues involved in stereo-vision-based algorithms, this paper proposes real-time monocular-vision-based techniques for simultaneous vehicle detection and inter-vehicle distance estimation, in which the performance and robustness of the system remain competitive, even for highly challenging benchmark datasets. This paper develops a collision warning system by detecting vehicles ahead and, by identifying safety distances to assist a distracted driver, prior to occurrence of an imminent crash. We introduce adaptive global Haar-like features for vehicle detection, tail-light segmentation, virtual symmetry detection, intervehicle distance estimation, as well as an efficient single-sensor multifeature fusion technique to enhance the accuracy and robustness of our algorithm. The proposed algorithm is able to detect vehicles ahead at both day or night and also for short-and long-range distances. Experimental results under various weather and lighting conditions (including sunny, rainy, foggy, or snowy) show that the proposed algorithm outperforms state-of-the-art algorithms.

show abstract

Global Haar-Like Features: A New Extension of Classic Haar Features for Efficient Face Detection in Noisy Images

Cited by 14 publications

References 17 publications

Prototyping An Autonomous Eye-Controlled System (AECS) using Raspberry-Pi on Wheelchairs

Prototyping An Autonomous Eye-Controlled System (AECS) using Raspberry-Pi on Wheelchairs

Body Gestures Recognition System to Control a Service Robot

Robust Vehicle Detection and Distance Estimation Under Challenging Lighting Conditions

Contact Info

Product

Resources

About