Segmentation free shared weight networks for automatic vehicle detection

被引：24

作者：

Gader, PD ^{[1
]}

Miramonti, JR ^{[1
]}

Won, Y ^{[1
]}

Coffield, P ^{[1
]}

机构：

[1] USAF,EGLIN AFB,FL 32542

来源：

NEURAL NETWORKS | 1995年 / 8卷 / 09期

关键词：

shared weight networks; multilayer perceptrons; automatic target detection; image algebra; Matched filters; minimum average correlation energy filters; backpropagation; vehicle detection;

D O I：

10.1016/0893-6080(95)00068-2

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The problem of detecting a vehicle of a certain type in an outdoor scene is investigated. A shared weight neural network is applied in correlation fashion to images of outdoor scenes. Two sets of images are considered. The first set consists of forward looking infrared images of tanks. The second set consists of visible images of cars in a parking lot. The goal in the first set of images is to detect the tanks. The goal in the second set of images is to detect the Chevrolet Blazers. The second problem is much more difficult than the first. The networks are compared to a minimum average correlation energy (MACE) matched filter approach. The shared weight network was found to be better at generalizing. Problems were found in trying to suppress spurious outputs on the background. As a result, an improved training algorithm was developed.

引用

页码：1457 / 1473

页数：17

共 38 条

[1] Arsenault, Hsu, Optical pattern recognition using circular-harmonic expansion, Applied Optics, 21, pp. 4016-4019, (1982)
[2] Casasent, Unified synthetic discriminant function computational formulation, Applied Optics, 23, 10, pp. 1620-1627, (1984)
[3] Casasent, Chang, Correlation synthetic discriminant functions, Applied Optics, 25, 14, pp. 2343-2350, (1984)
[4] Casasent, Ravichandran, Modified MACE filters for distortion invariant recognition of mobile target, Paper presented at the SPIE Airborne Reconnaissance XIII, (1989)
[5] Casasent, Ravichandran, Advanced distortion-invariant minimum average correlation energy (MACE) filters, Applied Optics, 31, 8, pp. 1109-1116, (1992)
[6] Casasent, Ravichandran, Bollapragada, Gaussian-minimum average correlation energy filters, Applied Optics, 30, 35, pp. 5176-5181, (1991)
[7] Casasent, Lyer, Ravichandran, Circular-harmonic function minimum average correlation energy filters, Applied Optics, 30, 35, pp. 5169-5175, (1992)
[8] Gader, Fuzzy morphological networks, First Midwest Electro-Technology Conference, (1992)
[9] Gader, Template generation for pattern classification, SPIE Image Algebra and Morphological Image Processing III, (1992)
[10] Gader, Forester, Gillies, Ganzberger, Vogt, Trenkle, A segmentation-free neural network classifier for machine-printed numeric fields, 5th United States Postal Service Advanced Technology Conference, (1992)

← 1 2 3 4 →