Photogrammetry in real estate cadaster

The geodesic method of the characteristic points’ coordinates measuring is the main method for urban cadastral works (including complex ones). Implementing digital aerial photography cameras, unmanned aerial vehicles and improving hardware and software systems for image processing enable achieving the necessary accuracy (10 cm in plan coordinates) when using the photogrammetric method. Stereo models and orthomosaics are the output products of the mentioned technology using for measurements. Due to the fact that at creating an orthomosaic, additional image conversion processes are required and they may cause the loss of accuracy and the presence of perspective distortions of high-altitude objects, orthomosaics cannot be used to determine the coordinates of characteristic points. It is proposed to use a stereo model, i.e. a three-dimensional high-precision image of the terrain, as a product for measuring characteristic points in cadastral works. The experiments’ results and the experience of production work proved that the accuracy of geodesic and stereophotogrammetric methods in the real estate cadaster are equal. At the same time, the mentioned method has some advantages

Monitoring of Corroded and Loosened Bolts in Steel Structures via Deep Learning and Hough Transforms

In this study, a regional convolutional neural network (RCNN)-based deep learning and Hough line transform (HLT) algorithm are applied to monitor corroded and loosened bolts in steel structures. The monitoring goals are to detect rusted bolts distinguished from non-corroded ones and also to estimate bolt-loosening angles of the identified bolts. The following approaches are performed to achieve the goals. Firstly, a RCNN-based autonomous bolt detection scheme is designed to identify corroded and clean bolts in a captured image. Secondly, a HLT-based image processing algorithm is designed to estimate rotational angles (i.e., bolt-loosening) of cropped bolts. Finally, the accuracy of the proposed framework is experimentally evaluated under various capture distances, perspective distortions, and light intensities. The lab-scale monitoring results indicate that the suggested method accurately acquires rusted bolts for images captured under perspective distortion angles less than 15° and light intensities larger than 63 lux.

Image completion via transformation and structural constraints

Image completion is an approach to fill a damaged region (hole) in an image. In this study, we adopt a novel method which can repair a target region with structural constraints in an architectural scene. An objective function that consists of three terms is proposed to solve the image completion problem. In color term, we compute a parameterized transformation model using detected plane parameters and measure the distance between the target patch and transformed source patch. This model helps to extend the patch search space and find an optimal solution. To improve the patch matching accuracy, we add a guide term that includes structure term and consistency term. The structure term encourages sampling patches along the structural direction, and the consistency term is used to maintain the texture consistency. Considering the color deviation between patches, we add a gradient term into a framework that can solve more challenging problems. Compared with previous methods, the proposed method has good performance in preserving global structure and reasonably estimating perspective distortions. Moreover, we obtain acceptable results in natural scenes. The experimental results illustrate that this novel method is a potential tool for image completion.

Blind First-Order Perspective Distortion Correction Using Parallel Convolutional Neural Networks

In this work, we present a network architecture with parallel convolutional neural networks (CNN) for removing perspective distortion in images. While other works generate corrected images through the use of generative adversarial networks or encoder-decoder networks, we propose a method wherein three CNNs are trained in parallel, to predict a certain element pair in the 3×3 transformation matrix, M^. The corrected image is produced by transforming the distorted input image using M^−1. The networks are trained from our generated distorted image dataset using KITTI images. Experimental results show promise in this approach, as our method is capable of correcting perspective distortions on images and outperforms other state-of-the-art methods. Our method also recovers the intended scale and proportion of the image, which is not observed in other works.

FDCNet: Frontend-Backend Fusion Dilated Network Through Channel-Attention Mechanism

Crowd counting has attracted much attention in computer vision owing to its fundamental contribution in public security. But due to occlusions, perspective distortions, scale variations, and background interference it faces a great challenge. In this paper we propose a novel model to count crowds, named FDCNet: frontend-backend fusion dilated network through channel-attention mechanism. It merges the frontend feature map with the backend feature map, achieving a fusion of various scale features without additional branches or extra subtasks. The fusion is fed into the channel-attention block to optimize the procedure and to conduct feature recalibration to use global and spatial information. Furthermore, we utilize dilated layers to obtain a high-quality density map, and the SSIM-based loss function is added to compare the local correlation between the estimated density map and the ground truth. Our FDCNet is verified in four common datasets and gets a brilliant estimation.

Would Gaze-Contingent Rendering Improve Depth Perception in Virtual and Augmented Reality?

Near distances are overestimated in virtual reality, and far distances are underestimated, but an explanation for these distortions remains elusive. One potential concern is that whilst the eye rotates to look at the virtual scene, the virtual cameras remain static. Could using eye-tracking to change the perspective of the virtual cameras as the eye rotates improve depth perception in virtual reality? This paper identifies 14 distinct perspective distortions that could in theory occur from keeping the virtual cameras fixed whilst the eye rotates in the context of near-eye displays. However, the impact of eye movements on the displayed image depends on the optical, rather than physical, distance of the display. Since the optical distance of most head-mounted displays is over 1m, most of these distortions will have only a negligible effect. The exception are ‘gaze-contingent disparities’, which will leave near virtual objects looking displaced from physical objects that are meant to be at the same distance in augmented reality.