Target Detection With Semisupervised Kernel Orthogonal Subspace Projection

Capobianco, Luca; Garzelli, Andrea; Camps‐Valls, Gustau

doi:10.1109/tgrs.2009.2020910

Cited by 57 publications

(22 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, good performance of the above approaches could be achieved only when the manually labeled samples are provided. To alleviate the tedious and unreliable manual annotation, some researchers adopted semi-supervised learning (SSL) methods to perform object detection [18,19], in which only a few labeled training samples were used to train detectors and then new samples in training set were added from unlabeled data. However, these methods still require a comparative number of manual labeled positive examples.…”

Section: Introductionmentioning

confidence: 99%

Negative Bootstrapping for Weakly Supervised Target Detection in Remote Sensing Images

Zhou

Zhang

Cheng

et al. 2015

2015 IEEE International Conference on Multimedia Big Data

View full text Add to dashboard Cite

When training a classifier in a traditional weakly supervised learning scheme, negative samples are obtained by randomly sampling. However, it may bring deterioration or fluctuation for the performance of the classifier during the iterative training process. Considering a classifier is inclined to misclassify negative examples which resemble positive ones, comprising these misclassified and informative negatives should be important for enhancing the effectiveness and robustness of the classifier. In this paper, we propose to integrate Negative Bootstrapping scheme into weakly supervised learning framework to achieve effective target detection in remote sensing images. Compared with traditional weakly supervised target detection schemes, this method mainly has three advantages. Firstly, our model training framework converges more stable and faster by selecting the most discriminative training samples. Secondly, on each iteration, we utilize the negative samples which are most easily misclassified to refine target detector, obtaining better performance. Thirdly, we employ a pre-trained convolutional neural network (CNN) model named Caffe to extract high-level features from RSIs, which carry more semantic meanings and hence yield effective image representation. Comprehensive evaluations on a high resolution airplane dataset and comparisons with state-of-the-art weakly supervised target detection approaches demonstrate the effectiveness and robustness of the proposed method.

show abstract

Section: Introductionmentioning

confidence: 99%

Negative Bootstrapping for Weakly Supervised Target Detection in Remote Sensing Images

Zhou

Zhang

Cheng

et al. 2015

2015 IEEE International Conference on Multimedia Big Data

View full text Add to dashboard Cite

show abstract

“…Essentially, we propose to deform the kernel using the graph Laplacian. This idea was originally presented in [11] for the inductive SVM and has been recently presented for kernel orthogonal subspace projection target detection [15]. Here, it is applied to the OC-SVM.…”

Section: Proposed Smentioning

confidence: 99%

“…For our experiments, class "Graminoid marsh" was selected as the target class, whereas the others are considered as outlier class. The interest in this class is motivated by its intrinsic complexity; it is underrepresented, and it can be confused with similar subclasses in the scene, as documented in [15]. …”

Section: ) Hyperspectral Crop Detectionmentioning

confidence: 99%

Semisupervised One-Class Support Vector Machines for Classification of Remote Sensing Data

Muñoz-Marí

Bovolo

Gómez-Chova

et al. 2010

IEEE Trans. Geosci. Remote Sensing

232

108

View full text Add to dashboard Cite

Abstract-This paper presents two semisupervised one-class support vector machine (OC-SVM) classifiers for remote sensing applications. In one-class image classification, one tries to detect pixels belonging to one of the classes in the image and reject the others. When few labeled pixels of only one class are available, obtaining a reliable classifier is a difficult task. In the particular case of SVM-based classifiers, this task is even harder because the free parameters of the model need to be finely adjusted, but no clear criterion can be adopted. In order to improve the OC-SVM classifier accuracy and alleviate the problem of free-parameter selection, the information provided by unlabeled samples present in the scene can be used. In this paper, we present two state-of-the-art algorithms for semisupervised one-class classification for remote sensing classification problems. The first proposed algorithm is based on modifying the OC-SVM kernel by modeling the data marginal distribution with the graph Laplacian built with both labeled and unlabeled samples.

show abstract

“…Thus, target detection is one of the important applications of HIS [5]. It can be viewed as a binary classification problem where pixels are labeled as target (target present) or background (target absent) based on their spectral characteristics [6]. The most common model assumes that the spectra are represented by unique spatially non-overlapping materials.…”

Section: Introductionmentioning

confidence: 99%