Multiple instance learning for landmine detection using ground penetrating radar

Manandhar, Achut; Morton, Kenneth D.; Collins, Leslie M.; Torrione, Peter A.

doi:10.1117/12.917947

Cited by 7 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, during classifier development and execution, other ad hoc techniques to aggregate decision metrics from sets of images are required (e.g., average the top three confidences as in Section IV-B). This process is necessary because the underlying problem being solved is not a simple pattern recognition problem but can instead be well represented as a multiple-instance learning (MIL) problem (e.g., [39], [40], [44]- [46]). Treating the learning problem in GPR as an MIL problem is a young but active area of research that is currently being explored.…”

Section: Discussionmentioning

confidence: 99%

“…As with many GPR-processing applications, both the training and testing procedures used here are beset by logistical problems necessitating ad hoc solutions, e.g., the selection of positive examples with smoothed energy, and aggregating over classifier confidences using order statistics. In actuality these problems can be seen as symptoms of the underlying multipleinstance nature of the target identification problem [39], [40]; see Section VI for further discussion.…”

Section: B Temporal Locations For Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Histograms of Oriented Gradients for Landmine Detection in Ground-Penetrating Radar Data

Torrione

Morton

Sakaguchi

et al. 2014

IEEE Trans. Geosci. Remote Sensing

159

100

View full text Add to dashboard Cite

Ground-penetrating radar (GPR) is a powerful and rapidly maturing technology for subsurface threat identification. However, sophisticated processing of GPR data is necessary to reduce false alarms due to naturally occurring subsurface clutter and soil distortions. Most currently fielded GPR-based landmine detection algorithms utilize feature extraction and statistical learning to develop robust classifiers capable of discriminating buried threats from inert subsurface structures. Analysis of these techniques indicates strong underlying similarities between efficient landmine detection algorithms and modern techniques for feature extraction in the computer vision literature. This paper explores the relationship between and application of one modern computer vision feature extraction technique, namely histogram of oriented gradients (HOG), to landmine detection in GPR data. The results presented indicate that HOG features provide a robust tool for target identification for both classification and prescreening and suggest that other techniques from computer vision might also be successfully applied to target detection in GPR data. Index Terms-Computer vision, edge histogram descriptors, ground-penetrating radar (GPR), histogram of oriented gradients (HOG), random forest.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: B Temporal Locations For Feature Extractionmentioning

confidence: 99%

Histograms of Oriented Gradients for Landmine Detection in Ground-Penetrating Radar Data

Torrione

Morton

Sakaguchi

et al. 2014

IEEE Trans. Geosci. Remote Sensing

159

100

View full text Add to dashboard Cite

show abstract

“…In [140,141], MIL classifiers detect buried landmines from ground-penetrating radar signals. When a detection occurs at a given GPS coordinate, measures are taken at various depths in the soil.…”

Section: Other Applicationsmentioning

confidence: 99%

Multiple instance learning: A survey of problem characteristics and applications

Carbonneau¹,

Cheplygina²,

Granger³

et al. 2018

Pattern Recognition

527

313

View full text Add to dashboard Cite

Multiple instance learning (MIL) is a form of weakly supervised learning where training instances are arranged in sets, called bags, and a label is provided for the entire bag. This formulation is gaining interest because it naturally fits various problems and allows to leverage weakly labeled data. Consequently, it has been used in diverse application fields such as computer vision and document classification. However, learning from bags raises important challenges that are unique to MIL. This paper provides a comprehensive survey of the characteristics which define and differentiate the types of MIL problems. Until now, these problem characteristics have not been formally identified and described. As a result, the variations in performance of MIL algorithms from one data set to another are difficult to explain. In this paper, MIL problem characteristics are grouped into four broad categories: the composition of the bags, the types of data distribution, the ambiguity of instance labels, and the task to be performed. Methods specialized to address each category are reviewed. Then, the extent to which these characteristics manifest themselves in key MIL application areas are described. Finally, experiments are conducted to compare the performance of 16 state-of-the-art MIL methods on selected problem characteristics. This paper provides insight on how the problem characteristics affect MIL algorithms, recommendations for future benchmarking and promising avenues for research.

show abstract

“…This method has previously been applied to landmine data using Edge Histogram Descriptors (EHD) features at different depths. 12 The problem with MIL in this application is that the constraints of the model break down as the size of the image region decreases. With small image regions, which are desirable in order to identify the specific parts of a landmine response, the assumptions for MIL no longer hold.…”

Section: Introductionmentioning

confidence: 99%

Application of image categorization methods for buried threat detection in GPR data

et al. 2013

Self Cite

View full text Add to dashboard Cite

Utilizing methods from the image processing and computer vision fields has led to advances in high resolution Ground Penetrating Radar (GPR) based threat detection. By analyzing 2-D slices of GPR data and applying various image processing algorithms, it is possible to discriminate between threat and non-threat objects. In initial attempts to utilize such approaches, object instance-matching algorithms were applied to GPR images, but only limited success was obtained when utilizing feature point methods to identify patches of data that displayed landmine-like characteristics. While the approach worked well under some conditions, the instance-matching method of classification was not designed to identify a type of class, only reproductions of a specific instance. In contrast, our current approach is focused on identifying methods that can account for within-class variations that result from changing target types and varying operating conditions that a GPR system regularly encounters. Image category recognition is an area of research that attempts to account for within class variation of objects within visual images. Instead of finding a reproduction of a particular known object within an image, algorithms for image categorization are designed to learn the qualities of images that contain an instance belonging to a known class. The results illustrate how image category recognition algorithms can be successfully applied to threat identification in GPR data.

show abstract

Multiple instance learning for landmine detection using ground penetrating radar

Cited by 7 publications

References 0 publications

Histograms of Oriented Gradients for Landmine Detection in Ground-Penetrating Radar Data

Histograms of Oriented Gradients for Landmine Detection in Ground-Penetrating Radar Data

Multiple instance learning: A survey of problem characteristics and applications

Application of image categorization methods for buried threat detection in GPR data

Contact Info

Product

Resources

About