Image retrieval performance can be improved by training a convolutional neural network (CNN) model with annotated data to facilitate accurate localization of target regions. However, obtaining sufficiently annotated data is expensive and impractical in real settings. It is challenging to achieve accurate localization of target regions in an unsupervised manner. To address this problem, we propose a new unsupervised image retrieval method named unsupervised target region localization (UTRL) descriptors. It can precisely locate target regions without supervisory information or learning. Our method contains three highlights: 1) we propose a novel zero-label transfer learning method to address the problem of co-localization in target regions. This enhances the potential localization ability of pretrained CNN models through a zero-label data-driven approach; 2) we propose a multiscale attention accumulation method to accurately extract distinguishable target features. It distinguishes the importance of features by using local Gaussian weights; and 3) we propose a simple yet effective method to reduce vector dimensionality, named twice-PCA-whitening (TPW), which reduces the performance degradation caused by feature compression. Notably, TPW is a robust and general method that can be widely applied to image retrieval tasks to improve retrieval performance. This work also facilitates the development of image retrieval based on short vector features. Extensive experiments on six popular benchmark datasets demonstrate that our method achieves about 7% greater mean average precision (mAP) compared to existing state-of-the-art unsupervised methods.

Recently, deep convolutional neural networks (DCNNs) have been widely adopted in scene classification for remote sensing (RS) images, yielding impressive results. One important assumption for training these models is that the image labels are free from errors. However, the mislabeling of certain images can severely degrade the performance of DCNN models. While existing methods offer some degree of mitigation, their effectiveness diminishes significantly in the presence of highly contaminated labels. This letter seeks to address this issue by proposing a robust scene classification model designed to learn from noisy labels. First, a semi-supervised approach is proposed, involving bootstrapping a model that uses a cleaned training set through coarse screening of potentially noisy labels. In particular, the model is trained using consistency regularization to enhance its robustness. Second, the remaining images with uncertain labels are further used to increase the number of images available for training. Here, dynamic thresholding is applied to detect and correct label errors. Lastly, the images and their labels from the two steps above are jointly used for final training. The proposed method is evaluated on three widely used RS datasets (AID, WHU-RS19, and UCMerced). The results demonstrate that the proposed approach outperforms competing methods, highlighting its efficacy in addressing label noise in RS datasets.

Enhancing the accuracy of dense classification with limited labeled data and abundant unlabeled data, known as semi-supervised semantic segmentation, is an essential task in vision comprehension. Due to the lack of annotation in unlabeled data, additional pseudo-supervised signals, typically pseudo-labeling, are required to improve the performance. Although effective, these methods fail to consider the internal representation of neural networks and the inherent class-imbalance in dense samples. In this work, we propose an information transfer theory, which establishes a theoretical relationship between shallow and deep representations. We further apply this theory at both the semantic and pixel levels, referred to as IIT-SP, to align different types of information. The proposed IIT-SP optimizes shallow representations to match the target representation required for segmentation. This limits the upper bound of deep representations to enhance segmentation performance. We also propose a momentum-based Cluster-State bar that updates class status online, along with a HardClassMix augmentation and a loss weighting technique to address class imbalance issues based on it. The effectiveness of the proposed method is demonstrated through comparative experiments on PASCAL VOC and Cityscapes benchmarks, where the proposed IIT-SP achieves state-of-the-art performance, reaching mIoU of 68.34% with only 2% labeled data on PASCAL VOC and mIoU of 64.20% with only 12.5% labeled data on Cityscapes.

Anomaly detection has been widely explored by training an out-of-distribution detector with only normal data for medical images. However, detecting local and subtle irregularities without prior knowledge of anomaly types brings challenges for lung CT-scan image anomaly detection. In this paper, we propose a self-supervised framework for learning representations of lung CT-scan images via both multi-scale cropping and simple masked attentive predicting, which is capable of constructing a powerful out-of-distribution detector. Firstly, we propose CropMixPaste, a self-supervised augmentation task for generating density shadow-like anomalies that encourage the model to detect local irregularities of lung CT-scan images. Then, we propose a self-supervised reconstruction block, named simple masked attentive predicting block (SMAPB), to better refine local features by predicting masked context information. Finally, the learned representations by self-supervised tasks are used to build an out-of-distribution detector. The results on real lung CT-scan datasets demonstrate the effectiveness and superiority of our proposed method compared with state-of-the-art methods.

In image retrieval, highlighting target object and reducing the influence of background noise remains challenging. To address this problem, we propose a novel weighting method that aggregates deep convolutional features based on filtering, called filtering on spatial channel weighting (FSCW) factors, to represent image contents, and utilize it for image retrieval. There are three main contributions of this study. First, the designed filter can effectively remove the influence of background noise. Second, we propose a new channel selection and spatial weighting method, which can accurately distinguish target object from the background noise. Finally, we designed a new channel weighting strategy to suppress intra-image visual burstiness. Experimental results on benchmark datasets demonstrate that the proposed method effectively enhances discriminative power and outperforms some existing state-of-the-art methods in terms of the mAP metric. Furthermore, the proposed method is superior to some existing algorithms in distinguishing background noise and target object.

Center loss is widely used as a supervision tool in deep learning method. However, the center loss also has some shortcomings, the most important of which is that it must be combined with softmax loss to run well. In this article, we sum up five shortcomings of center loss and solve all of them by proposing a dual distance center loss (DDCL). Compared with center loss, DDCL can run without the combination of softmax to supervise training the model. In addition, we verify the inconsistency between the proposed DDCL and softmax loss in the feature space. To be specifically, we add the Pearson distance on the basis of the Euclidean distance to the same center, which makes all features of the same class be confined to the intersection of a hypersphere and a hypercone in the feature space, strengthens the intraclass compactness of the center loss, and enhances the generalization ability of center loss. Moreover, by designing a Euclidean distance threshold between all center pairs, we not only strengthen the interclass separability of center loss, but also make the center loss (or DDCL) works well without the combination of softmax loss. We verify the effectiveness of DDCL in four datasets, two of which are widely used in the field of vehicle re-identification named VeRi-776 dataset and VehicleID dataset, and two other datasets are widely used in the field of person re-identification named Market1501 dataset and MSMT17 dataset. The experimental results of the proposed DDCL exceed that of the softmax loss in all the four datasets, indicating that our proposed method not only can run without the combination of softmax, but also has a high accuracy.

Non-local self similarity (NSS) is a powerful prior of natural images for image denoising. Most of existing denoising methods employ similar patches, which is a patch-level NSS prior. In this paper, we take one step forward by introducing a pixel-level NSS prior, i.e., searching similar pixels across a non-local region. This is motivated by the fact that finding closely similar pixels is more feasible than similar patches in natural images, which can be used to enhance image denoising performance. With the introduced pixel-level NSS prior, we propose an accurate noise level estimation method, and then develop a blind image denoising method based on the lifting Haar transform and Wiener filtering techniques. Experiments on benchmark datasets demonstrate that, the proposed method achieves much better performance than previous non-deep methods, and is still competitive with existing state-of-the-art deep learning based methods on real-world image denoising. The code is publicly available athttps://github.com/njusthyk1972/NLH.

Human visual perception has a close relationship with the HSV color space, which can be represented as a cylinder. The question of how visual features are extracted using such an attribute is important. In this paper, a new feature descriptor; namely, a color volume histogram, is proposed for image representation and content-based image retrieval. It converts a color image from RGB color space to HSV color space and then uniformly quantizes it into 72 bins of color cues and 32 bins of edge cues. Finally, color volumes are used to represent the image content. The proposed algorithm is extensively tested on two Corel datasets containing 15[Formula: see text]000 natural images. These image retrieval experiments show that the color volume histogram has the power to describe color, texture, shape and spatial features and performs significantly better than the local binary pattern histogram and multi-texton histogram approaches.

Foreground and background cues can assist humans in quickly understanding visual scenes. In computer vision, however, it is difficult to detect salient objects when they touch the image boundary. Hence, detecting salient objects robustly under such circumstances without sacrificing precision and recall can be challenging. In this paper, we propose a novel model for salient region detection, namely, the foreground-center-background (FCB) saliency model. Its main highlights as follows. First, we use regional color volume as the foreground, together with perceptually uniform color differences within regions to detect salient regions. This can highlight salient objects robustly, even when they touched the image boundary, without greatly sacrificing precision and recall. Second, we employ center saliency to detect salient regions together with foreground and background cues, which improves saliency detection performance. Finally, we propose a novel and simple yet efficient method that combines foreground, center, and background saliency. Experimental validation with three well-known benchmark data sets indicates that the FCB model outperforms several state-of-the-art methods in terms of precision, recall, F-measure, and particularly, the mean absolute error. Salient regions are brighter than those of some existing state-of-the-art methods.