Salient Object Detection via Recursive Sparse Representation

A New Road Damage Detection Baseline with Attention Learning

Automated detection of road damage (ADRD) is a challenging topic in road maintenance. It focuses on automatically detecting road damage and assessing severity by deep learning. Because of the sparse distribution of characteristic pixels, it is more challenging than object detection. Although some public datasets provide a database for the development of ADRD, their amounts of data and the standard of classification cannot meet network training and feature learning. With the aim of solving this problem, this work publishes a new road damage dataset named CNRDD, which is labeled according to the latest evaluation standard for highway technical conditions in China (JTG5210-2018). The dataset is collected by professional onboard cameras and is manually labeled in eight categories with three different degrees (mild, moderate and severe), which can effectively help promote research of automated detection of road damage. At the same time, a novel baseline with attention fusion and normalization is proposed to evaluate and analyze the published dataset. It explicitly leverages edge detection cues to guide attention for salient regions and suppresses the weights of non-salient features by attention normalization, which can alleviate the interference of sparse pixel distribution on damage detection. Experimental results demonstrate that the proposed baseline significantly outperforms most existing methods on the existing RDD2020 dataset and the newly released CNRDD dataset. Further, the CNRDD dataset is proved more robust, as its high damage density and professional classification are more conducive to promote the development of ADRD.

Detection and classification of burnt skin via sparse representation of signals by over-redundant dictionaries

Skin burns in color images must be accurately detected and classified according to burn degree in order to assist clinicians during diagnosis and early treatment. Especially in emergency cases in which clinical experience might not be available to conduct a thorough examination with high accuracy, an automated assessment may benefit patient outcomes. In this work, detection and classification of burnt areas are performed by using the sparse representation of feature vectors by over-redundant dictionaries. Feature vectors are extracted from image patches so that each patch is assigned to a class representing a burn degree. Using color and texture information as features, detection and classification achieved 95.65% sensitivity and 94.02% precision. Experiments used two methods to build dictionaries for burn severity classes to apply to observed skin regions: (1) direct collection of feature vectors from patches in various images and locations and (2) collection of feature vectors followed by dictionary learning accompanied by K-singular value decomposition.

Cirrus Detection Based on RPCA and Fractal Dictionary Learning in Infrared imagery

In earth observation systems, especially in the detection of small and weak targets, the detection and recognition of long-distance infrared targets plays a vital role in the military and civil fields. However, there are a large number of high radiation areas on the earth’s surface, in which cirrus clouds, as high radiation areas or abnormal objects, will interfere with the military early warning system. In order to improve the performance of the system and the accuracy of small target detection, the method proposed in this paper uses the suppression of the cirrus cloud as an auxiliary means of small target detection. An infrared image was modeled and decomposed into thin parts such as the cirrus cloud, noise and clutter, and low-order background parts. In order to describe the cirrus cloud more accurately, robust principal component analysis (RPCA) was used to get the sparse components of the cirrus cloud, and only the sparse components of infrared image were studied. The texture of the cirrus cloud was found to have fractal characteristics, and a random fractal based infrared image signal component dictionary was constructed. The k-cluster singular value decomposition (KSVD) dictionary was used to train the sparse representation of sparse components to detect cirrus clouds. Through the simulation test, it was found that the algorithm proposed in this paper performed better on the the receiver operating characteristic (ROC) curve and Precision-Recall (PR) curve, had higher accuracy rate under the same recall rate, and its F-measure value and Intersection-over-Union (IOU) value were greater than other algorithms, which shows that it has better detection effect.

Fusion of Multispectral and Panchromatic Images via Spatial Weighted Neighbor Embedding

Fusing the panchromatic (PAN) image and low spatial-resolution multispectral (LR MS) images is an effective technology for generating high spatial-resolution MS (HR MS) images. Some image-fusion methods inspired by neighbor embedding (NE) are proposed and produce competitive results. These methods generally adopt Euclidean distance to determinate the neighbors. However, closer Euclidean distance is not equal to greater similarity in spatial structure. In this paper, we propose a spatial weighted neighbor embedding (SWNE) approach for PAN and MS image fusion, by exploring the similar manifold structures existing in the observed LR MS images to those of HR MS images. In SWNE, the spatial neighbors of the LR patch are found first. Second, the weights of these neighbors are estimated by the alternative direction multiplier method (ADMM), in which the neighbors and their weights are determined simultaneously. Finally, the HR patches are reconstructed by the sum of HR patches corresponding to the LR patches multiplying with their weights. Due to the introduction of spatial structures in objective function, outlier patches can be eliminated effectively by ADMM. Compared with other methods based on NE, more reasonable neighbor patches and their weights are estimated simultaneously. Some experiments are conducted on datasets collected by QuickBird and Geoeye-1 satellites to validate the effectiveness of SWNE, and the results demonstrate a better performance of SWNE in spatial and spectral information preservation.

An Improved Boosting Learning Saliency Method for Built-Up Areas Extraction in Sentinel-2 Images

Built-up areas extraction from satellite images is an important aspect of urban planning and land use; however, this remains a challenging task when using optical satellite images. Existing methods may be limited because of the complex background. In this paper, an improved boosting learning saliency method for built-up area extraction from Sentinel-2 images is proposed. First, the optimal band combination for extracting such areas from Sentinel-2 data is determined; then, a coarse saliency map is generated, based on multiple cues and the geodesic weighted Bayesian (GWB) model, that provides training samples for a strong model; a refined saliency map is subsequently obtained using the strong model. Furthermore, cuboid cellular automata (CCA) is used to integrate multiscale saliency maps for improving the refined saliency map. Then, coarse and refined saliency maps are synthesized to create a final saliency map. Finally, the fractional-order Darwinian particle swarm optimization algorithm (FODPSO) is employed to extract the built-up areas from the final saliency result. Cities in five different types of ecosystems in China (desert, coastal, riverside, valley, and plain) are used to evaluate the proposed method. Analyses of results and comparative analyses with other methods suggest that the proposed method is robust, with good accuracy.