Improving RGB-D SLAM in dynamic environments using semantic aided segmentation

Summary Most conventional simultaneous localization and mapping (SLAM) approaches assume the working environment to be static. In a highly dynamic environment, this assumption divulges the impediments of a SLAM algorithm that lack modules that distinctively attend to dynamic objects despite the inclusion of optimization techniques. This work exploits such environments and reduces the effects of dynamic objects in a SLAM algorithm by separating features belonging to dynamic objects and static background using a generated binary mask image. While the features belonging to the static region are used for performing SLAM, the features belonging to non-static segments are reused instead of being eliminated. The approach employs deep neural network or DNN-based object detection module to obtain bounding boxes and then generates a lower resolution binary mask image using depth-first search algorithm over the detected semantics, characterizing the segmentation of the foreground from the static background. In addition, the features belonging to dynamic objects are tracked into consecutive frames to obtain better masking consistency. The proposed approach is tested on both publicly available dataset as well as self-collected dataset, which includes both indoor and outdoor environments. The experimental results show that the removal of features belonging to dynamic objects for a SLAM algorithm can significantly improve the overall output in a dynamic scene.

Sparsity-Based Recovery of Three-Dimensional Photoacoustic Images from Compressed Single-Shot Optical Detection

Photoacoustic (PA) imaging combines optical excitation with ultrasonic detection to achieve high-resolution imaging of biological samples. A high-energy pulsed laser is often used for imaging at multi-centimeter depths in tissue. These lasers typically have a low pulse repetition rate, so to acquire images in real-time, only one pulse of the laser can be used per image. This single pulse necessitates the use of many individual detectors and receive electronics to adequately record the resulting acoustic waves and form an image. Such requirements make many PA imaging systems both costly and complex. This investigation proposes and models a method of volumetric PA imaging using a state-of-the-art compressed sensing approach to achieve real-time acquisition of the initial pressure distribution (IPD) at a reduced level of cost and complexity. In particular, a single exposure of an optical image sensor is used to capture an entire Fabry–Pérot interferometric acoustic sensor. Time resolved encoding as achieved through spatial sweeping with a galvanometer. This optical system further makes use of a random binary mask to set a predetermined subset of pixels to zero, thus enabling recovery of the time-resolved signals. The Two-Step Iterative Shrinking and Thresholding algorithm is used to reconstruct the IPD, harnessing the sparsity naturally occurring in the IPD as well as the additional structure provided by the binary mask. We conduct experiments on simulated data and analyze the performance of our new approach.

Method of Increasing the Availability of Video Information of Aerial Monitoring in the Airspace of a City

The study of the factors that form the threats to the violation of the properties of the availability and integrity of aeromonitoring video information in the system of prevention and elimination of crisis situations. The direction of increasing the availability of video information based on the use of compression technology for encoding video data has been substantiated. It is shown that to eliminate the drawback associated with lowering the lower boundary of the differential polyadic space, it is necessary for the perforation technology to additionally take into account the binary mask of burst elements of the upper and lower levels. The main conceptual components of the image compression method to increase the availability of video information, based on the coding of composite numbers with a mask in a differential perforated polyadic space, have been developed. The main results of a comparative assessment of the basic component of information availability in aeromonitoring systems are presented.

Morphological characteristics and distribution of dolines in Slovenia, a study of a lidar-based doline map of Slovenia

Dolines are small to intermediate enclosed depressions and are the most numerous karst feature in Slovenia. They are circular in plan form and vary in diameter from a few metres to over a kilometre. They are developed in limestone, dolomite, carbonate breccia and conglomerate and occupy different geomorphic settings. They were formed by various processes like dissolution, collapse, suffosion and transformation of caves to surface features by denudation. Publicly accessible lidar data, provided by a nationwide laser scanning project of Slovenia, was used for this study. To catalogue the dolines, we manually label a fraction of the digital elevation model (DEM) with a binary mask indicating if the area is a doline or not. We then train a slightly modified u-net, a type of machine learning algorithm, on the labelled territory. Using the trained algorithm, we infer the binary mask on the entire DEM. We convert the resulting mask into an ESRI Shapefile and manually verify the results. We note that the training and inference are error prone on types of relief that were less common in the training set (e.g., the relatively uncommon collapse dolines). We believe manual verification mitigates most of these errors, so the resulting map is a good basis for the doline study. We have made our georeferenced catalogue of dolines available at https://dolines.org/ (Mihevc & Mihevc 2021). Dolines are found in most of the karst areas, except mountains where they were eroded by glacial action or covered by glacial deposits. We detected 471,192 dolines and divided them into three genetic types. Most abundant are solution dolines (470,325). The average doline is 9 m deep, has a diameter of 42 m and a volume of 14,098 m3. The density of dolines on levelled surfaces can be as high as 500/ per km2. They are absent from the floors of poljes and steeper slopes, and are less abundant on sloping surfaces. We have identified 314 dolines to be of collapse origin. The mean depth of collapse dolines is 49 m, and 20 of them are deeper than 100 m. The mean volume is 1.2 million m3, with the largest having a volume of 11.6 million m3. Most of the collapse dolines can be found close to ponors or springs or corridors where large underground rivers flow. We have detected 553 suffosion dolines formed by suffosion of sediments in blind valleys or on poljes. This basic data set for dolines enables further study and comparison of dolines with the geology and topography of the karst.

Stamps extraction using local adaptive k- means and ISODATA algorithms

<span>One of the main difficulties facing the certified documents documentary archiving system is checking the stamps system, but, that stamps may be contains complex background and surrounded by unwanted data. Therefore, the main objective of this paper is to isolate background and to remove noise that may be surrounded stamp. Our proposed method comprises of four phases, firstly, we apply k-means algorithm for clustering stamp image into a number of clusters and merged them using ISODATA algorithm. Secondly, we compute mean and standard deviation for each remaining cluster to isolate background cluster from stamp cluster. Thirdly, a region growing algorithm is applied to segment the image and then choosing the connected region to produce a binary mask for the stamp area. Finally, the binary mask is combined with the original image to extract the stamp regions. The results indicate that the number of clusters can be determined dynamically and the largest cluster that has minimum standard deviation (i.e., always the largest cluster is the background cluster). Also, show that the binary mask can be established from more than one segment to cover are all stamp’s disconnected pieces and it can be useful to remove the noise appear with stamp region.</span>

The video processing features research in computer systems and special purpose networks

For a typical low complexity video sequence, the weight of each P-frame in the stream is approximately three times smaller than the I-frame weight. However, taking into account the number of P-frames in the group, they make the main contribution to the total video data amount. Therefore, the possibility of upgrading coding methods for P-frames is considered on preliminary blocks' type identification with the subsequent formation of block code structures. As the correlation coefficient between adjacent frames increases, the compression ratio of the differential-represented frame's binary mask increases. The compression ratio of the differential-represented frame's binary mask varies from 3 to 21 depending on the correlation coefficient between adjacent frames. The most preferable method for constructing the compact representation technology of the binary masks of frames represented in a differential form is the approach. This is based on the identification and description of the lengths of one-dimensional binary series. A binary series is a consecutive binary elements sequence with the same value. In this case, sequences of identical binary elements are replaced by their lengths.

Saliency Cuts: Salient Region Extraction based on Local Adaptive Thresholding for Image Information Recognition of the Visually Impaired

In recent years, there has been an increased scope for assistive software and technologies, which help the visually impaired to perceive and recognize natural scene images. In this article, we propose a novel saliency cuts approach using local adaptive thresholding to obtain four regions from a given saliency map. The saliency cuts approach is an effective tool for salient object detection. First, we produce four regions for image segmentation using a saliency map as an input image and applying an automatic threshold operation. Second, the four regions are used to initialize an iterative version of the Grab Cut algorithm and to produce a robust and high-quality binary mask with a full resolution. Lastly, based on the binary mask and extracted salient object, outer boundaries and internal edges are detected by Canny edge detection method. Extensive experiments demonstrate that the proposed method correctly detects and extracts the main contents of the image sequences for delivering visually salient information to the visually impaired people compared to the results of existing salient object segmentation algorithms

An Effective Framework for Speech and Music Segregation

Speech and music segregation from a single channel is a challenging task due to background interference and intermingled signals of voice and music channels. It is of immense importance due to its utility in wide range of applications such as music information retrieval, singer identification, lyrics recognition and alignment. This paper presents an effective method for speech and music segregation. Considering the repeating nature of music, we first detect the local repeating structures in the signal using a locally defined window for each segment. After detecting the repeating structure, we extract them and perform separation using a soft time-frequency mask. We apply an ideal binary mask to enhance the speech and music intelligibility. We evaluated the proposed method on the mixtures set at -5 dB, 0 dB, 5 dB from Multimedia Information Retrieval-1000 clips (MIR-1K) dataset. Experimental results demonstrate that the proposed method for speech and music segregation outperforms the existing state-of-the-art methods in terms of Global-Normalized-Signal-to-Distortion Ratio (GNSDR) values