In this paper, we give an overview of image matching techniques for various vision-based navigation systems: stereo vision, structure from motion and map-based approach. Focused on map-based approach, which generally uses feature-based matching for localization, and based on our early developed system, a performance analysis has been carried out and three major problems have been identified: being vulnerable to illumination changes, drastic viewpoint changes and good percentage of mismatches. By introducing ASIFT into the system, the major improvement takes place on the epoch with large viewpoint changes. In order to deal with mismatches that are unable to be removed by RANSAC, we propose to use cross-correlation information to evaluate the quality of homography model and help select the proper one. The conducted experiments have proved that such an approach can reduce the chances of mismatches being included by RANSAC and final positioning accuracy can be improved.
Position information of individual nodes is very useful in implementing functions such as routing, querying and many applications. Researchers have proposed several location techniques but only a few relative location algorithms for mobile wireless sensor networks (WSNs). Although mobility would appear to make localization more difficult, this paper introduces a distributed 3 dimension relative localization algorithm for the mobile WSNs. This approach is an effort in finding a solution to positioning problem for highly dynamic nodes deployed randomly in a complex environment and the 3 dimension coordinate provided can be satisfied with great applications. The distributed coordinate transformation algorithm takes lower convergence times in the establishment of the global coordinate system.
1. Machine (especially deep) learning algorithms are changing the way wildlife imagery is processed. They dramatically speed up the time to detect, count classify animals and their behaviours. Yet, we currently lack a systematic literature survey on its use in wildlife imagery.2. Through a literature survey (a ‘rapid’ review) and bibliometric mapping, we explored its use across: 1) species (vertebrates), 2) image types (e.g., camera traps, or drones), 3) study locations, 4) alternative machine learning algorithms, 5) outcomes (e.g., recognition, classification, or tracking), 6) reporting quality and openness, 7) author affiliation, and 8) publication journal types.3. Typically, studies have focused on single large charismatic or iconic mammalian species and used neural networks (i.e., deep learning). Additional taxa or alternative machine learning algorithms were rarely used, with limited sharing of code. There were considerable gaps, and therefore there is a great promise for deep learning to transform behavioural detection, classification, and tracking of wildlife.4. Much of the published research and focus on animals came from India, China, Australia, or the USA. There were relatively few collaborations across countries. Given the power of machine learning, we recommend increasing collaboration and sharing approaches to utilise increasing amounts of wildlife imagery more rapidly and transform and improve understanding of wildlife behaviour and conservation.5. Our survey augmented with bibliometric analyses provide valuable signposts for future studies to resolve and address shortcomings, gaps, and biases.
Image matching is an important step in the process of power inspection image stitching and 3D reconstruction, which directly affects the effect of power line image stitching and 3D reconstruction, and is widely used in power line inspection. Traditional power inspection image matching applications are mostly implemented on X86 architecture, but ARM architecture has gradually entered the server field in recent years, showing high efficiency and energy-saving advantages, so it is necessary to transplant power inspection image matching applications to the server platform of ARM architecture. Since the application of image matching is very computation-based, in the process of porting the algorithm to the ARM platform, this paper adopts the Gaussian pyramid algorithm to accelerate the application of image matching through the search method from coarse to fine.
Power equipment is an important part of the power system and the focus of power system operation and maintenance. Infrared anomaly detection technology is an effective means to detect abnormalities of power equipment because of its safety, simplicity and intuitiveness. Through training the YOLOv3 network by infrared images collected in the field, this work can achieve real-time detection of power equipment and fault points on the Jetson Nano, and determines which areas of the power equipment are abnormal. The trained YOLOv3 model is tested. The mAP value of the model is 34.63%, the recall rate is 21%, and the temperature anomaly area and power equipment could be marked. The running time on the Jetson Nano was 0.7-0.9 s (the recognition time was less than 1s), which satisfies the requirements for power equipment testing.
Society faces a severe environmental challenge posed by the rapid advance of technology scaling. The high cost in manufacturing energy, materials, and disposal is worrisome with the increasing number of smartphones. To mitigate the impact of future devices, the authors propose a design for reuse model in which obsolete devices will be reused for a class of applications that can be satisfied with older, less reliable technology. In particular, the authors find a good match between the reuse of smartphones and educational applications. The experiments indicate that the resource requirements of educational applications can be satisfied by repurposed smartphones. The key challenge is the design of software that can adapt to extreme heterogeneity of devices. To this end, the authors explore smartphone evolutions and characterize different types of heterogeneities among different generations of smartphones. The authors propose insights to aid establishing a sustainable model of designing mobile applications for phone reuse.
The application of Wireless Sensor Networks (WSNs) is restrained by their often-limited lifetime. A sensor node's lifetime is fundamentally linked to the volume of data that it senses, processes and reports. Spatial correlation between sensor nodes is an inherent phenomenon to WSNs, induced by redundant nodes which report duplicated information. In this paper, we report on the design of a distributed sampling scheme referred to as the `Virtual Sampling Scheme (VSS)'. VSS is formed from two components: an algorithm for forming virtual clusters and a distributed sampling method. VSS primarily utilizes redundancy of sensor nodes to get only a subset to sense the environment at any one time. Sensor nodes that are not sensing the environment are in a low-power sleep state, thus conserving energy. Furthermore, VSS balances the energy consumption amongst nodes by using a round robin method.
The application of trajectory classification to automatically detect movement types of unknown trajectories has been receiving increasing research attention in areas such as video surveillance, traffic management and location-based services. Existing research applies classic geometric shape-based classification approaches to classify trajectories by utilizing the geometric characteristics of movement to fulfill this task. However, this approach is limited to the geographic context of trajectory data. Classification methods based on movement parameters can overcome this problem but the accuracy of classification depends heavily on selecting appropriate movement features from trajectories. This research proposes an efficient trajectory classification model based on two types of complexity measures as new features for classifying movements: (1) the geometric complexity measures of trajectories based on Fractal Dimensions, and (2) structural complexity measures of movement parameters based on Approximate Entropy (ApEn). We suggest that ApEn, which provides complexity information about the subtle changes that occur in the structure of sequential movement parameters of trajectories, and Fractal Dimensions, which provide the overall description of geometric complexity, can be used together to improve the accuracy in trajectory classification. The feasibility of this proposed classification model is tested with 800 GPS trajectories that were shared and manually tagged with four movement types by Internet users on the website Openstreemap.org. The overall 85.4% average accuracy of prediction demonstrates the applicability of this classification model. By improving the quality of trajectory classification, the proposed approach in this research will benefit many applications of trajectory data analysis and mining.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
'/%3e%3c/defs%3e%3cpath fill='%23012169' d='M0 0h10080v5040H0z'/%3e%3cpath stroke='%23fff' stroke-width='.6' d='m0 0 6 3m0-3L0 3' clip-path='url(%23b)' transform='scale(840)'/%3e%3cpath stroke='%23e4002b' stroke-width='.4' d='m0 0 6 3m0-3L0 3' clip-path='url(%23c)' transform='scale(840)'/%3e%3cpath stroke='%23fff' stroke-width='840' d='M2520 0v2520M0 1260h5040'/%3e%3cpath stroke='%23e4002b' stroke-width='504' d='M2520 0v2520M0 1260h5040'/%3e%3cg fill='%23fff'%3e%3cuse xlink:href='%23d' x='2520' y='3780'/%3e%3cuse xlink:href='%23a' x='7560' y='4200'/%3e%3cuse xlink:href='%23a' x='6300' y='2205'/%3e%3cuse xlink:href='%23a' x='7560' y='840'/%3e%3cuse xlink:href='%23a' x='8680' y='1869'/%3e%3cuse xlink:href='%23e' x='8064' y='2730'/%3e%3c/g%3e%3c/svg%3e)