An automatic detection system for buried explosive hazards in FL-LWIR and FL-GPR data

Stone, Kevin; Keller, James M.; Anderson, Derek T.; Barclay, David

doi:10.1117/12.920288

Cited by 23 publications

(18 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our pre-screener is an ensemble of trainable size-contrast filters with mean-shift clustering 7,8 . The goal of a pre-screener Figure 1: Visual illustration of the processing stages put forth in this article.…”

Section: Methodsmentioning

confidence: 99%

“…Using a non-edge class in conjunction with an edge threshold, pixels whose maximum response is less than the edge threshold for any of the four directional edge masks is assigned to the non-edge class. In Stone et al 7 , improvements were made to the method in 9 to make the EHD more robust (e.g., less sensitive to noise). For full details of the improved approach, see 7 .…”

Section: Soft Edge Histogram Descriptormentioning

confidence: 99%

See 1 more Smart Citation

Investigation of context, soft spatial, and spatial frequency domain features for buried explosive hazard detection in FL-LWIR

et al. 2014

View full text Add to dashboard Cite

It is well-known that a pattern recognition system is only as good as the features it is built upon. In the fields of image processing and computer vision, we have numerous spatial domain and spatial-frequency domain features to extract characteristics of imagery according to its color, shape and texture. However, these approaches extract information across a local neighborhood, or region of interest, which for target detection contains both object(s) of interest and background (surrounding context). A goal of this research is to filter out as much task irrelevant information as possible, e.g., tire tracks, surface texture, etc., to allow a system to place more emphasis on image features in spatial regions that likely belong to the object(s) of interest. Herein, we outline a procedure coined soft feature extraction to refine the focus of spatial domain features. This idea is demonstrated in the context of an explosive hazards detection system using forward looking infrared imagery. We also investigate different ways to spatially contextualize and calculate mathematical features from shearlet filtered candidate image chips. Furthermore, we investigate localization strategies in relation to different ways of grouping image features to reduce the false alarm rate. Performance is explored in the context of receiver operating characteristic curves on data from a U.S. Army test site that contains multiple target and clutter types, burial depths, and times of day.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Soft Edge Histogram Descriptormentioning

confidence: 99%

Investigation of context, soft spatial, and spatial frequency domain features for buried explosive hazard detection in FL-LWIR

et al. 2014

View full text Add to dashboard Cite

show abstract

“…The amount of incoming data to process is very large, not to mention the issues of image to world registration, aka georeferencing. In previous works [12][13], we have addressed these issues by developing a fast prescreening algorithm consisting of an ensemble of size-contrast filters which generates hit locations in world coordinates from an image stream captured by a camera mounted on a moving vehicle. This process uses a synthetic stereo approach, described in [14], to allow accurate transformation of 2D image coordinates to 3D world coordinates at standoff distances in excess of 30 meters.…”

Section: Notes On Application Conditions and Preprocessingmentioning

confidence: 99%

Convolutional neural network approach for buried target recognition in FL-LWIR imagery

Stone

Keller

2014

SPIE Proceedings

Self Cite

View full text Add to dashboard Cite

A convolutional neural network (CNN) approach to recognition of buried explosive hazards in forward-looking long-wave infrared (FL-LWIR) imagery is presented. The convolutional filters in the first layer of the network are learned in the frequency domain, making enforcement of zero-phase and zero-dc response characteristics much easier. The spatial domain representations of the filters are forced to have unit l2 norm, and penalty terms are added to the online gradient descent update to encourage orthonormality among the convolutional filters, as well smooth first and second order derivatives in the spatial domain. The impact of these modifications on the generalization performance of the CNN model is investigated. The CNN approach is compared to a second recognition algorithm utilizing shearlet and log-gabor decomposition of the image coupled with cell-structured feature extraction and support vector machine classification. Results are presented for multiple FL-LWIR data sets recently collected from US Army test sites. These data sets include vehicle position information allowing accurate transformation between image and world coordinates and realistic evaluation of detection and false alarm rates.

show abstract

“…An inertial navigation system (INS) was mounted next to the camera and the time at which each image was captured was recorded. This allowed precise georeferencing using the dense 3D scene reconstruction technique described in 8 . Experiments are reported using receiver operating characteristic (ROC) curves on an arid U.S. Army test site that contains multiple target and clutter types, burial depths, and times of day.…”

Section: Fl-ehd Prescreenermentioning

confidence: 99%

“…The point is, there are still fewer candidate locations to sort through with a prescreener than to look at all images and all possible translations, rotations and scalings. Our group has put forth a number of FLIR prescreeners to date 7,8 . However, it is always of interest to explore different prescreeners for purposes such as higher detection rates, lower FARs and/or fusing different hit lists from different prescreeners.…”

Section: Introductionmentioning

confidence: 99%

Extended adaptive mutation operator for training an explosive hazard detection prescreener in forward looking infrared imagery

2015

View full text Add to dashboard Cite

A big challenge with forward looking (FL), versus downward looking, sensors mounted on a ground vehicle for explosive hazard detection (EHD) is they "see everything", on and off road. Even if a technology such as road detection is used, we still have to find and subsequently discriminate targets versus clutter on the road and often road side. When designing an automatic detection system for FL-EHD, we typically make use of a prescreener to identify regions of interest (ROI) instead of searching for targets in an inefficient brute force fashion by extracting complicated features and running expensive classifiers at every possible translation, rotation and scale. In this article, we explore the role of genetic algorithms (GAs), specifically with respect to a new adaptive mutation operator, for learning the parameters of a FL-EHD prescreener in FL infrared (FLIR) imagery. The proposed extended adaptive mutation (eAM) algorithm is driven by fitness similarities in the chromosome population. Currently, our prescreener consists of many free parameters that are empirically chosen by a researcher. The parameters are learned herein using the proposed optimization technique and the performance of the system is measured using receiver operating characteristic (ROC) curves on data obtained from a U.S. Army test site that includes a variety of target types buried at varying depths and from different times of day. The proposed technique is also applied to numerous synthetic fitness landscapes to further assess the effectiveness of the eAM algorithm. Results show that the new adaptive mutation technique converges faster to a better solution than a GA with fixed mutation.

show abstract

An automatic detection system for buried explosive hazards in FL-LWIR and FL-GPR data

Cited by 23 publications

References 0 publications

Investigation of context, soft spatial, and spatial frequency domain features for buried explosive hazard detection in FL-LWIR

Investigation of context, soft spatial, and spatial frequency domain features for buried explosive hazard detection in FL-LWIR

Convolutional neural network approach for buried target recognition in FL-LWIR imagery

Extended adaptive mutation operator for training an explosive hazard detection prescreener in forward looking infrared imagery

Contact Info

Product

Resources

About