Semantic segmentation methods based on deep learning considerably improve the segmentation performance of remote sensing images. However, with the extensive application of high-resolution remote sensing images, additional details introduce considerable interference to the learning process for classification, thereby diminishing the accuracy of segmentation and resulting in blurry object boundaries. To address this problem, this study designed Random-Walk-SegNet (RWSNet), a semantic segmentation network based on SegNet combined with random walk. First, SegNet is used as the basic architecture with the sliding window strategy that optimizes the network output to improve the continuity and smoothness of segmentation. Second, seed regions of the random walk are selected in accordance with the classification output of SegNet. Third, the weights of the undirected graph edge are determined by fusing the gradient of the original image and probability map of SegNet. Finally, random walk is implemented on the entire image, thus reducing edge blur and realizing high-performance semantic segmentation of remote sensing images. In comparison with mainstream and other improved methods, the proposed network has lower complexity but better performance, and the algorithm is state-of-the-art and robust.

Image registration is a basic but essential step for remote sensing image processing, and finding stable features in multitemporal images is one of the most considerable challenges in the field. The main shape contours of artificial objects (e.g., roads, buildings, farmlands, and airports) can be generally described as a group of line segments, which are stable features, even in images with evident background changes (e.g., images taken before and after a disaster). In this study, a registration method that uses line segments and their intersections is proposed for multitemporal remote sensing images. First, line segments are extracted in image pyramids to unify the scales of the reference image and the test image. Then, a line descriptor based on the gradient distribution of local areas is constructed, and the segments are matched in image pyramids. Lastly, triplets of intersections of matching lines are selected to estimate affine transformation between two images. Additional corresponding intersections are provided based on the estimated transformation, and an iterative process is adopted to remove outliers. The performance of the proposed method is tested on a variety of optical remote sensing image pairs, including synthetic and real data. Compared with existing methods, our method can provide more accurate registration results, even in images with significant background changes.

Dual cameras with visible-thermal multispectral pairs provide both visual and thermal appearance, thereby enabling detecting pedestrians around the clock in various conditions and applications, including autonomous driving and intelligent transportation systems. However, due to the greatly varying real-world scenarios, the performance of a detector trained on a source dataset might change dramatically when evaluated on another dataset. A large amount of training data is often necessary to guarantee the detection performance in a new scenario. Typically, human annotators need to conduct the data labeling work, which is time-consuming, labor-intensive and unscalable. To overcome the problem, we propose a novel unsupervised transfer learning framework for multispectral pedestrian detection, which adapts a multispectral pedestrian detector to the target domain based on pseudo training labels. In particular, auxiliary detectors are utilized and different label fusion strategies are introduced according to the estimated environmental illumination level. Intermediate domain images are generated by translating the source images to mimic the target ones, acting as a better starting point for the parameter update of the pedestrian detector. The experimental results on the KAIST and FLIR ADAS datasets demonstrate that the proposed method achieves new state-of-the-art performance without any manual training annotations on the target data.

Person re-identification (re-id) aims to retrieve images of same identities across different camera views. Resolution mismatch occurs due to varying distances between person of interest and cameras, this significantly degrades the performance of re-id in real world scenarios. Most of the existing approaches resolve the re-id task as low resolution problem in which a low resolution query image is searched in a high resolution images gallery. Several approaches apply image super resolution techniques to produce high resolution images but ignore the multiple resolutions of gallery images which is a better realistic scenario. In this paper, we introduce channel correlations to improve the learning of features from the degraded data. In addition, to overcome the problem of multiple resolutions we propose a Resolution based Feature Distillation (RFD) approach. Such an approach learns resolution invariant features by filtering the resolution related features from the final feature vectors that are used to compute the distance matrix. We tested the proposed approach on two synthetically created datasets and on one original multi resolution dataset with real degradation. Our approach improves the performance when multiple resolutions occur in the gallery and have comparable results in case of single resolution (low resolution re-id).

Recently, pedestrian detection using visible-thermal pairs plays a key role in around-the-clock applications, such as public surveillance and autonomous driving. However, the performance of a well-trained pedestrian detector may drop significantly when it is applied to a new scenario. Normally, to achieve a good performance on the new scenario, manual annotation of the dataset is necessary, while it is costly and unscalable. In this work, an unsupervised transfer learning framework is proposed for visible-thermal pedestrian detection tasks. Given well-trained detectors from a source dataset, the proposed framework utilizes an iterative process to generate and fuse training labels automatically, with the help of two auxiliary single-modality detectors (visible and thermal). To achieve label fusion, the knowledge of daytime and nighttime is adopted to assign priorities to labels according to their illumination, which improves the quality of generated training labels. After each iteration, the existing detectors are updated using new training labels. Experimental results demonstrate that the proposed method obtains state-of-the-art performance without any manual training labels on the target dataset.

In this paper, we propose a new cooperative method that improves the accuracy of Turn Movement Count (TMC) under challenging conditions by introducing contextual observations from the surrounding areas. The proposed method focuses on the correct identification of the movements in conditions where current methods have difficulties. Existing vision-based TMC systems are limited under heavy traffic conditions. The main problems for most existing methods are occlusions between vehicles that prevent the correct detection and tracking of the vehicles through the entire intersection and the assessment of the vehicle’s entry and exit points, incorrectly assigning the movement. The proposed method intends to overcome this incapability by sharing information with other observation systems located at neighboring intersections. Shared information is used in a cooperative scheme to infer the missing data, thereby improving the assessment that would otherwise not be counted or miscounted. Experimental evaluation of the system shows a clear improvement over related reference methods.