Detecting driver drowsiness using feature-level fusion and user-specific classification

详细信息    查看全文

• 作者：Jaeik Jo ; Sung Joo Lee ; Kang Ryoung Park ; Ig-Jae Kim ; Jaihie Kim
• 关键词：ASM ; active shape model ; CCA ; canonical correlation analysis ; ECD ; eye closure duration ; EER ; equal error rate ; ESD-Value ; eyelid state detection value ; FEC ; frequency of eye closure ; GLCM ; grey-level co-occurrence matrix ; GRBF ; Gaussian radial basis function ; LBP ; local binary pattern ; LDA ; linear discriminant analysis ; NHTSA ; national highway traffic safety administration ; PCA ; principal component analysis ; PERCLOS ; percentage of eye closure ; PHOGs ; pyramid histogram of oriented gradients ; ROI
• 刊名：Expert Systems with Applications
• 出版年：March, 2014
• 年：2014
• 卷：41
• 期：4_part_P1
• 页码：1139-1152
• 全文大小：3928 K

文摘

Accurate classification of eye state is a prerequisite for preventing automobile accidents due to driver drowsiness. Previous methods of classification, based on features extracted for a single eye, are vulnerable to eye localization errors and visual obstructions, and most use a fixed threshold for classification, irrespective of variations in the driver鈥檚 eye shape and texture. To address these deficiencies, we propose a new method for eye state classification that combines three innovations: (1) extraction and fusion of features from both eyes, (2) initialization of driver-specific thresholds to account for differences in eye shape and texture, and (3) modeling of driver-specific blinking patterns for normal (non-drowsy) driving. Experimental results show that the proposed method achieves significant improvements in detection accuracy.