Active learning for detecting a spectrally variable subject in color infrared imagery

Foschi, Patricia G.; Liu, Huan

doi:10.1016/j.patrec.2004.06.004

Cited by 5 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The presence of this exotic weed has disrupted navigation and recreational uses of waterways, clogged irrigation intake trenches, and caused reservoir-pumping problems [15]. The Egeria invasion has also displaced native flora and probably affected native fauna.…”

Section: Egeria Detectionmentioning

confidence: 99%

“…Classifying Egeria in scan-digitized CIR imagery presents a challenging problem due to a number of variable and unfavorable conditions [15]. These include changes in imaging conditions (e.g., film exposure, vignetting, scanning anomalies), problems associated with water-related subjects (e.g., turbidity, sun glint, surface reflectance due to wind), and other environmental changes (e.g., exposure of Egeria at extremely low tide, shadows falling upon the water, algal cover over Egeria).…”

Section: Egeria Detectionmentioning

confidence: 99%

“…Manually obtaining labels for massive unlabeled data is not only time consuming but also unreliable. Experts can only process a small portion of the unlabeled images in a given period of time [15]. One goal of our work is to reduce human involvement in the labeling process by applying some learning methodology to automate this process of labeling for a large number of images.…”

Section: An Active Learning Approachmentioning

confidence: 99%

See 2 more Smart Citations

An Active Learning Approach to Egeria Densa Detection in Digital Imagery

Kantardzic¹,

Zurada²

2011

Next Generation of Data-Mining Applications

View full text Add to dashboard Cite

This chapter focuses on the development of an active learning approach to an image mining problem for detecting Egeria densa (a Brazilian waterweed) in digital imagery. An effective way of automatic image classification is to employ learning systems. However, due to a large number of images, it is often impractical to manually create labeled data for supervised learning. On the other hand, classification systems generally require labeled data to carry out learning. In order to strike a balance between the difficulty of obtaining labeled images and the need for labeled data, we explore an active learning approach to image mining. The goal is to minimize the task of expert labeling of images: if labeling is necessary, only those important parts of an image will be presented to experts for labeling. The critical issues are: (1) how to determine what should be presented to experts; (2) how to minimize the number of those parts for labeling; and (3) after a small number of labeled instances are available, how to effectively learn a classifier and apply it to new images. We propose to use ensemble methods for active learning in Egeria detection. Our approach is to use the combined classifications of the ensemble of classifiers to reduce the number of uncertain instances in the image classification process and thus achieve reduced expert involvement in image labeling. We demonstrate the effectiveness of our proposed system via experiments using a real-world application of Egeria detection. Practical concerns in image mining using active learning are also addressed and discussed.2

show abstract

Section: Egeria Detectionmentioning

confidence: 99%

Section: Egeria Detectionmentioning

confidence: 99%

Section: An Active Learning Approachmentioning

confidence: 99%

See 1 more Smart Citation