A Novel Side Face Contour Extraction Algorithm for Driving Fatigue Statue Recognition

Lei, Jianmei; Han, Qingwen; Chen, Lei; Lai, Zhida; Zeng, Lingqiu; Liu, Xin

doi:10.1109/access.2017.2686424

Cited by 11 publications

(9 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcome the non-ideal conditions, the main image preprocessing technique is the color space modeling. In [13], Lei et al proposed a new color space model based on the combination of YIQ and YUV color spaces and used multi-threshold decision criteria for facial region recognition. Their method can enhance the facial skin extraction and eliminate the downside of the complexity background.…”

Section: Related Workmentioning

confidence: 99%

“…The feature extraction shall be greatly simplified if the luminance of all images are similar for DCNNs. To this end, we follow the Gray World's color equalization method [13] to implement the process. The process is illustrated as follows: i) Calculate the mean value of the RGB components of the image, marked as R avg , G avg , and B avg , respectively.…”

Section: ) Lm Streammentioning

confidence: 99%

“…The component I of YIQ space is closely related to human skin color. Previous study [13] gives the criterion rang setting of I ∈ [20,90] for skin color recognition. We believe that it is an important feature for DCNN extracting more facial features than the background features.…”

Section: ) Yiq Streammentioning

confidence: 99%

“…The RGB stream is used for capturing the characteristic of original input RGB images. The LM stream is used for preprocessing the input RGB images based on the method of Gray World's color equalization [13]. By doing so, we can improve the illuminance balance of non-ideal facial images influenced by unbalanced ambient illumination.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Ensemble Learning for Facial Age Estimation Within Non-Ideal Facial Imagery

Qian

Huang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Human facial age estimation has been widely used in many computer vision applications, including security surveillance, forensics, biometrics, human-computer interaction (HCI), and so on. We propose a facial age estimation method oriented to non-ideal facial imagery. The method consists of image preprocessing, feature extraction, and age predication. First, we preprocess non-ideal input images in RGB stream, luminance modified (LM) stream, and YIQ stream. Then, we leverage the deep convolutional neural networks (DCNNs) to extract the feature of images preprocessed in each stream. To reduce the training data volume and training complexity, we adopt the transfer learning to build the DCNN structure. With the extracted feature, the weak classifier equipped at every stream is designed to obtain a weak classification prediction of the age range. Moreover, in order to generate estimation, we use the ensemble learning to fuse the three weak classifiers. We design an integrated strategy algorithm based on the combination of voting method and weighted average method. The simulation results show that our proposed algorithm can improve the an exact match (AEM) and an error of one age category (AEO) by 4.75% and 6.75% compared with the best AEM and AEO of the three weak classifiers. Furthermore, in comparison with the unweighted average method, our proposed algorithm can improve the AEM and AEO by 8.68% and 12.79%, respectively. INDEX TERMS Facial age estimation, transfer learning, ensemble learning, deep convolutional neural network, image preprocessing.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: ) Lm Streammentioning

confidence: 99%

Section: ) Yiq Streammentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Ensemble Learning for Facial Age Estimation Within Non-Ideal Facial Imagery

Qian

Huang

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Images suffer from the illumination and pose variations Gheorghita Ghinea, Rajkumar Kannan, And Suresh Kannaiyan [8] proposed a system which showed invariant to illumination and pose variationswith the help of image gradients orientation. Jianmei Lei, Qingwen Han, Lei Chen, Zhida Lai, Lingqiu Zeng, And Xin Liu [9] here another color space displaying multi-threshold decision criteria used to enhance facial feature extraction and a methodology is used to remove and reduce the noise and other impacts. Tests demonstrate in the system that the side face contour lines can be successfully extracted using the proposed algorithm.…”

Section: Literature Surveymentioning

confidence: 99%

A Two Level Face Recognition Algorithm For Efficient Attendance Marking

Bhaskar¹,

Shwetha²,

Rashmi³

et al. 2017

IOSR JECE

View full text Add to dashboard Cite

A face recognition is ascending as an exuberant research region traversing many controls like image processing, design acknowledgment, pattern recognition, PC vision and neural systems. Face recognition innovation has different business applications and another business. Here this paper introduces an automated complex face recognition based on the combination of w-Quartette and v-HOG which prepares the face picture for recognition and interpret the extraction information. As a piece of the proposed calculations, Filters of Wavelets have been executed utilizing multi-resolution approaches because of their adaptability, reasonableness, and normality for face acknowledgment. The proposed framework enhances the execution of a current face recognition framework by reducing out the occlusions, human intervention, marking and entry of attendance.

show abstract