Presentation 1993/10/15
Vehicle Classification based on Sound and Image Sensor Data Fusion
Tadashi Nakanishi, Akio Shio, Kenichiro Ishii,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a new method for classifying laterally moving vehicles from image and sound sequences.In our sensor data fusion technique,1)velocity,length,and shape of the silhouette 2) shape of the sound power spectrum are used to classify the vehicles.The new method has the following characteristics:1)based on the spatio-temporal image analysis,we can easily spot the sound sequences exactly derived from the target vehicle.2)by integrating characteristics extracted from image and sound data,vehicles including moter cycles,bicycles and passenger cars carrying laggage on their roof are classified correctly.Our experiments with real data confirm the effectiveness of the method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) sound / image / fusion / spatio-temporal image / power spectrum / vehicle classification
Paper # PRU93-73,HC93-47
Date of Issue

Conference Information
Committee PRU
Conference Date 1993/10/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Understanding (PRU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Vehicle Classification based on Sound and Image Sensor Data Fusion
Sub Title (in English)
Keyword(1) sound
Keyword(2) image
Keyword(3) fusion
Keyword(4) spatio-temporal image
Keyword(5) power spectrum
Keyword(6) vehicle classification
1st Author's Name Tadashi Nakanishi
1st Author's Affiliation NTT Human Interface Laboratries()
2nd Author's Name Akio Shio
2nd Author's Affiliation NTT Human Interface Laboratries
3rd Author's Name Kenichiro Ishii
3rd Author's Affiliation NTT Human Interface Laboratries
Date 1993/10/15
Paper # PRU93-73,HC93-47
Volume (vol) vol.93
Number (no) 269
Page pp.pp.-
#Pages 6
Date of Issue