Gabor-based dynamic representation for human fatigue monitoring in facial image sequences

Fan, Xiao; Sun, Yeneng; Yin, Baocai; Xiuming, Guo

doi:10.1016/j.patrec.2009.08.014

Cited by 40 publications

(18 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the above table 1, we can draw the conclusion that the correct rate of fatigue detection of the method proposed by [7] is closer to that of our proposed method, but the error detected rate is much higher than that of ours. From analysis, it is easy to know the reason is that the method proposed by [7] just used the single image to carry fatigue detection only, so it will classify all the Pseudo-fatigue…”

Section: Methodssupporting

confidence: 50%

“…Correct detection rate Error detection rate Our method 99.2% 0.02% Method proposed by [7] 98.6% 1.28%…”

Section: Methodsmentioning

confidence: 97%

“…And then, the mined fatigue sequential patterns are used to carry fatigue detection to all kinds of video sequences. We compared it with the method proposed by [7]. The results are show in table 1, in which the items used are calculated as follows:…”

Section: Algorithm Performance Analysismentioning

confidence: 99%

“…So there are some researchers proposed a fatigue detection method based on the whole facial features. The most representative methods include Gabor and Adaboost based method [7]、 dynamic Bayes network based method [8] and lasting time of eye closure and yawning patterns based method [9]. But in these methods, only a single facial image is used to judge the fatigue of drivers and the relationship between the former or the behind frames are not considered, thus, the detect rate of these methods are not very high also.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Gabor-based Facial Image Sequential Pattern Mining for Human Fatigue Monitoring

Wang¹,

Yang²,

Jiang³

et al. 2011

IJACT

View full text Add to dashboard Cite

Fatigue is the main factor to cause traffic accidents. Based on survey of existed fatigue detection methods, a new method for extracting characteristics of Human face based on Gabor translation is proposed in this paper, and based on which, by combing with the frequent patterns mining, a new fatigue detection method for vehicle drivers based on the Human face image sequences is proposed. Simulation results show that the newly proposed algorithm has better fatigue detection performance for identification of vehicle drivers than the existed method based on single face image.sequential-pattern can be generated by joining s 1 with s 2 , i.e., by extending the sequence s 1 with the last item in s 2 . Since the added item (i.e., the last item in s 2 ) will be a separate element in s 1 if it is a separate element in s 2 , or a part of the last element of s 1 otherwise, then, it is necessary to add the item in s 2 both as a part of an itemset and a separate element in s 1 when joining L k-1 with L k-1 .Step2(Prune Phase): Deleting all these candidate k-fatigue-sequential-patterns that have a contiguous (k-1)-sequence whose support count is less than the given support threshold min_support.sequences into fatigue models. But in our method, seldom will the pseudo-fatigue sequences be classified into fatigue models. Thus, the error detection rate of our method will be very low. ConclusionA new method for fatigue detection based on Gabor translation and human facial image sequence is proposed in this paper, in which, the Gabor translation is used to extract the feature sequences of the human facial image sequences first, and then, the frequent patterns mining algorithm is used to mine the patterns of fatigue facial image sequences, finally, during the recognition stage, the classification algorithm will be used for fatigue detection of the Human face image sequences. Simulation results show that the newly proposed algorithm has good fatigue detection performance.

show abstract

Section: Methodssupporting

confidence: 50%

“…Correct detection rate Error detection rate Our method 99.2% 0.02% Method proposed by [7] 98.6% 1.28%…”

Section: Methodsmentioning

confidence: 97%

Section: Algorithm Performance Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Gabor-based Facial Image Sequential Pattern Mining for Human Fatigue Monitoring

Wang¹,

Yang²,

Jiang³

et al. 2011

IJACT

View full text Add to dashboard Cite

show abstract

“…The driving fatigue detection methods have the following: driver's physiological signal based methods [5][6][7], behavior based methods, operation state of vehicle based method [8][9][10], and facial expressions based methods [11,12]. Physiological signal based methods requires the sensor in contact with the driver's body and detail requirements on the ways and points of contact, it is unsuitable for universal application of normal driving.…”

Section: Introductionmentioning

confidence: 99%

Recognition and Applications of Emotion Detection in Driving Fatigue

Hongzhe¹,

Gongzhang²

2015

TOAUTOCJ

View full text Add to dashboard Cite

For the driver at the long-term driving or at the poor mental state, the emotion based driving fatigue recognition algorithm is proposed, and we use the algorithm to effectively prevent the continuous driving of the driver in the state of fatigue, so as to avoid the accident. We sample the facial images of the driver and establish their feature space, and compare them with the fatigue training samples, achieve in real time the driving category attributes of the driver. According to the driving category attributes, we judge whether the driver is in the state of fatigue so as to appropriately intervene in the behavior of the driver.

show abstract

Driver Inattention Monitoring System for Intelligent Vehicles

Dong¹,

Hu²

2012

Encyclopedia of Sustainability Science and Technology

View full text Add to dashboard Cite

Gabor-based dynamic representation for human fatigue monitoring in facial image sequences

Cited by 40 publications

References 20 publications

Gabor-based Facial Image Sequential Pattern Mining for Human Fatigue Monitoring

Gabor-based Facial Image Sequential Pattern Mining for Human Fatigue Monitoring

Recognition and Applications of Emotion Detection in Driving Fatigue

Driver Inattention Monitoring System for Intelligent Vehicles

Contact Info

Product

Resources

About