Vision-Based Automated Recognition and 3D Localization Framework for Tower Cranes Using Far-Field Cameras.

Tower cranes can cover most of the area of a construction site, which brings significant safety risks, including potential collisions with other entities. To address these issues, it is necessary to obtain accurate and real-time information on the orientation and location of tower cranes and hooks. As a non-invasive sensing method, computer vision-based (CVB) technology is widely applied on construction sites for object detection and three-dimensional (3D) localization. However, most existing methods mainly address the localization on the construction ground plane or rely on specific viewpoints and positions. To address these issues, this study proposes a framework for the real-time recognition and localization of tower cranes and hooks using monocular far-field cameras. The framework consists of four steps: far-field camera autocalibration using feature matching and horizon-line detection, deep learning-based segmentation of tower cranes, geometric feature reconstruction of tower cranes, and 3D localization estimation. The pose estimation of tower cranes using monocular far-field cameras with arbitrary views is the main contribution of this paper. To evaluate the proposed framework, a series of comprehensive experiments were conducted on construction sites in different scenarios and compared with ground-truth data obtained by sensors. The experimental results show that the proposed framework achieves high precision in both crane jib orientation estimation and hook position estimation, thereby contributing to the development of safety management and productivity analysis.

Simultaneous Real-Time Three-Dimensional Localization and FRET Measurement of Two Distinct Particles.

Many biological processes employ mechanisms involving the locations and interactions of multiple components. Given that most biological processes occur in three dimensions, the simultaneous measurement of three-dimensional locations and interactions is necessary. However, the simultaneous three-dimensional precise localization and measurement of interactions in real time remains challenging. Here, we report a new microscopy technique to localize two spectrally distinct particles in three dimensions with an accuracy (2.35σ) of tens of nanometers with an exposure time of 100 ms and to measure their real-time interactions using fluorescence resonance energy transfer (FRET) simultaneously. Using this microscope, we tracked two distinct vesicles containing t-SNAREs or v-SNARE in three dimensions and observed FRET simultaneously during single-vesicle fusion in real time, revealing the nanoscale motion and interactions of single vesicles in vesicle fusion. Thus, this study demonstrates that our microscope can provide detailed information about real-time three-dimensional nanoscale locations, motion, and interactions in biological processes.

Monocular Pedestrian 3D Localization for Social Distance Monitoring.

Social distancing protocols have been highly recommended by the World Health Organization (WHO) to curb the spread of COVID-19. However, one major challenge to enforcing social distancing in public areas is how to perceive people in three dimensions. This paper proposes an innovative pedestrian 3D localization method using monocular images combined with terrestrial point clouds. In the proposed approach, camera calibration is achieved based on the correspondences between 2D image points and 3D world points. The vertical coordinates of the ground plane where pedestrians stand are extracted from the point clouds. Then, using the assumption that the pedestrian is always perpendicular to the ground, the 3D coordinates of the pedestrian's feet and head are calculated iteratively using collinear equations. This allows the three-dimensional localization and height determination of pedestrians using monocular cameras, which are widely distributed in many major cities. The performance of the proposed method was evaluated using two different datasets. Experimental results show that the pedestrian localization error of the proposed approach was less than one meter within tens of meters and performed better than other localization techniques. The proposed approach uses simple and efficient calculations, obtains accurate location, and can be used to implement social distancing rules. Moreover, since the proposed approach also generates accurate height values, exclusionary schemes to social distancing protocols, particularly the parent-child exemption, can be introduced in the framework.

3DLRA: An RFID 3D Indoor Localization Method Based on Deep Learning.

As the core supporting technology of the Internet of Things, Radio Frequency Identification (RFID) technology is rapidly popularized in the fields of intelligent transportation, logistics management, industrial automation, and the like, and has great development potential due to its fast and efficient data collection ability. RFID technology is widely used in the field of indoor localization, in which three-dimensional location can obtain more real and specific target location information. Aiming at the existing three-dimensional location scheme based on RFID, this paper proposes a new three-dimensional localization method based on deep learning: combining RFID absolute location with relative location, analyzing the variation characteristics of the received signal strength (RSSI) and Phase, further mining data characteristics by deep learning, and applying the method to the smart library scene. The experimental results show that the method has a higher location accuracy and better system stability.

Three-Dimensional Empirical AoA Localization Technique for Indoor Applications.

Small and pervasive devices have been increasingly used to identify and track objects automatically. Consequently, several low-cost localization schemes have been proposed in the literature based on angle of arrival (AoA), time difference of arrival (TDoA), received signal strength indicator (RSSI) or their combinations. In this paper, we propose a three-dimensional empirical AoA localization (TDEAL) technique for battery-powered devices. The proposed technique processes the AoA measurements at fixed reader nodes to estimate the locations of the tags. The proposed technique provides localization accuracy that mitigates non-linear empirical errors in AoA measurements. We utilize two omni-directional antenna arrays at each fixed reader node to estimate the location vector. With multiple location estimations from different fixed reader nodes, each estimated location is assigned a weight that is inversely proportional to the AoA phase-difference error. Furthermore, the actual AoA parabolic formula of the location is approximated to a cone to simplify the location calculation process. The proposed localization technique has a low hardware cost, low computational requirements, and precise location estimates. Based on the performance evaluation, significant location accuracy is achieved by TDEAL; where, for instance, an average error margin of less than 13 cm is achieved using 10 readers in an area of 10 × 10 m2. TDEAL can be utilized to provide reference points when integrated with a relative (e.g., inertial navigation systems) localization systems.

Small lung lesions invisible under fluoroscopy are located accurately by three-dimensional localization technique on chest wall surface and performed bronchoscopy procedures to increase diagnostic yields.

BACKGROUND: Nowadays, small peripheral pulmonary lesions (PPLs) are frequently detected and the prognosis of lung cancer depends on the early diagnosis. Because of the high fee and requiring specialized training, many advanced techniques are not available in many developing countries and rural districts. METHODS: Three sets of opaque soft copper wires visible under the fluoroscopy (Flu) in the Flu-flexible bronchoscopy (FB) group (n = 24), which determined the three planes of the lesion, were respectively placed firmly on the surface of the chest wall with adhesive tape on the chest wall. The FB tip was advanced into the bronchus toward the crosspoint of the three perpendicular planes under Flu with careful rotation of a C-arm unit. Then the specimen were harvested focusing around the crosspoint for pathologic diagnosis. The rapid on-site evaluation (ROSE) procedure was also performed. The average Flu time during FB procedures were recorded and diagnostic accuracy rates in the Flu-FB group were compared with the other group guided by radial endobronchial ultrasound (R-EBUS) (n = 23). RESULTS: The location of the core point of the lesion, whether it was visible or not under the fluoroscopy could be recognized by three-dimensional localization technique. The accuracy rates of diagnostic yields were 62.5% in the Flu-FB group, and was similar as 65.2% in the R-EBUS group (P > 0.05). However, in the Flu-FB group, there was a decreasing tendency on accurate diagnosis rates of lower lobe (LL) lesions when comparing with non-LL lesions (3/8 = 37.5% vs 12/16 = 75%, P = 0.091) while in the R-EBUS group it was similar (9/12 = 75% vs 6/11 = 54.6%, P = 0.278). In the Flu-FB group, fluoroscopy time was negatively correlated with the lesion length (r = -0.613, P = 0.001), however, there was no significant difference between the lesions invisible or not (5.83 ± 1.45 min vs 7.67 ± 2.02 min, P = 0.116) under the fluoroscopy, as well as no significant difference among SPN, mGGO and GGO (6.12 ± 2.05 min, 7.25 ± 1.33 min and 7.80 ± 2.02 min, P > 0.05). CONCLUSIONS: Small PPL whether it is visible or not under fluoroscopy can be located accurately by our three-dimensional localization technique on chest wall surface and performed bronchoscopy procedures to increase diagnostic yields. It is more convenient, economical and reliable with the similar diagnostic yields than R-EBUS guided method. TRIAL REGISTRATION: Current Controlled Trials ChiCTR-DDD-16009715 . The date of registration: 3rd Nov, 2016. Retrospectively registered.

Imaging and positioning through scattering media with double-helix point spread function engineering.

Significance: Double-helix point spread function (DH-PSF) microscopy has been developed for three-dimensional (3D) localization and imaging at super-resolution but usually in environments with no or weak scattering. To date, super-resolution imaging through turbid media has not been reported. Aim: We aim to explore the potential of DH-PSF microscopy in the imaging and localization of targets in scattering environments for improved 3D localization accuracy and imaging quality. Approach: The conventional DH-PSF method was modified to accommodate the scanning strategy combined with a deconvolution algorithm. The localization of a fluorescent microsphere is determined by the center of the corresponding double spot, and the image is reconstructed from the scanned data by deconvoluting the DH-PSF. Results: The resolution, i.e., the localization accuracy, was calibrated to 13 nm in the transverse plane and 51 nm in the axial direction. Penetration thickness could reach an optical thickness (OT) of 5. Proof-of-concept imaging and the 3D localization of fluorescent microspheres through an eggshell membrane and an inner epidermal membrane of an onion are presented to demonstrate the super-resolution and optical sectioning capabilities. Conclusions: Modified DH-PSF microscopy can image and localize targets buried in scattering media using super-resolution. Combining fluorescent dyes, nanoparticles, and quantum dots, among other fluorescent probes, the proposed method may provide a simple solution for visualizing deeper and clearer in/through scattering media, making in situ super-resolution microscopy possible for various demanding applications.

Percutaneous vertebroplasty guided by preoperative computed tomography measurements.

BACKGROUND: Percutaneous vertebroplasty (PVP) is now widely performed to treat painful vertebral compression fractures. Previous researches have reported numerous advantages. However, it rarely reported that how to determine the feasibility of the unilateral or bilateral approach and how to decide the puncture angle, the skin insertion site before the procedure. The aim of this study was to discuss the feasibility of PVP using unilateral pedicular approach by the three-dimensional positioning of computed tomography (CT) image. MATERIALS AND METHODS: Under fluoroscopic guidance, 108 patients with 115 diseased vertebral bodies underwent PVP. The study was divided in two groups. Group A, fifty patients with 52 vertebrae received PVP without using preoperative CT measurements and puncture simulation. Group B, 58 patients with 63 vertebrae received PVP using preoperative CT measurements and puncture simulation. The skin needle entry point and puncture angle of the transverse plane and sagittal plane were determined by the software of PACS on preoperative CT image. The choice of unilateral or bilateral pedicular approach was decided based on the CT image before the procedure. PVP was carried out according to the measurement result above. The average time for a single vertebra operation, the success rate of single puncture and complications was evaluated and compared between Group A and Group B. RESULTS: In Group A, technical success of unilateral PVP was 63.5% (33/52 vertebrae), and 92% (58/63 vertebrae) in Group B. The average time of operation in Groups A and B were (37.5 ± 5.5) and (28.5 ± 5.5) min, respectively. There was a significant difference in the time of single-vertebra operation and the success rates of unilateral PVP between Groups A and B. No serious complications developed during the followup period. CONCLUSIONS: The CT three-dimensional positioning measurement for PVP can increase the success rate of unilateral PVP.

Initial application of MRI and ultrasound three-dimensional matrix positioning technique for transperineal targeted prostate biopsy in patients with previous negative prostate biopsy / 第二军医大学学报

Objective To validate the value of MRI and ultrasound three-dimensional matrix positioning technique in transperineal targeted prostate biopsy so as to improve the positive rate of prostate biopsy. Methods A total of 15 patients received transperineal targeted prostate biopsy using MRI and ultrasound three-dimensional matrix positioning technique + traditional systemic prostate biopsy from Feb. 2015 to Jun. 2015 in Changhai Hospital, Second Military Medical University. The 15 patients met the following criteria:(1) at least one negative prostate biopsy; (2) had the indications for performing re-biopsy; and (3) MRI showed suspected sites of prostate cancer (the PI-RADS scores were 4-5). Before biopsy, all the MRI images were reconstructed and the three-dimensional coordinates(X, Y, Z axis) were confirmed. After general anesthesia, systemic prostate biopsy was performed under the guidance of transrectal ultrasound. Then the transrectal ultrasound-guided biopsy was carried out according to the location in previous MRI images, which were located by the three-dimensional coordinates (X, Y, Z axis). Results Fourteen (93. 3%) of the 15 patients were finally confirmed with prostate cancer, with 13 (86. 7%) being positive only in targeted biopsy, and 6 (40. 0%) being positive only in systemic biopsy. There was only 1 (6. 7%) whose targeted biopsy result was negative and systemic biopsy result was positive, and there were 8 cases (53. 3%) with positive targeted biopsy but negative systemic biopsy. Only 1 case (6. 7%) was negative in both targeted biopsy and systemic prostate biopsy. There was significant difference in positive rates between the targeted biopsy and systemic biopsy (P=0. 002). Conclusion Transperineal targeted prostate biopsy with MRI and ultrasound three-dimensional matrix positioning technique can accurately locate and target puncture the suspected site of prostate cancer in MRI images, and it has a higher diagnostic efficiency especially for the sites which are relatively easy to be missed by systemic prostate biopsy, but its performance still needs further large sample studies.

Magnetic Detection Electrical Impedance Tomography and its Application in Three Dimensional Localization of the Meridian / 中国医学物理学杂志

Objective: The Meridian Theory is an important component and the theory basis of Traditional Chinese Medicine. Three dimensional localization is the key to research of the Meridian Theory. Imaging is a scientific and practical method for the three dimensional localization of the meridian. Methods: The methods and actuality of the intrinsic quality and localization of meridians were first introduced. Then the methods and disadvantages of electrical impedance tomography (EIT) were also re-viewed. Based on these, the advantages and disadvantages of various methods applying to localization of meridian were analy-sised. Results: A new technology, magnetic detection electrical impedance tomography (MD-EIT) has obvious advantages com-paring to other EIT methods. It can fit comfortably on the dimentional localization of the meridian. Conclusions: MD-EIT can be applied not only to locate the meridians, but also to heart and lung detection and cancer diagnosis. There is a wide range of appli-cations for MD-EIT in medicine, It has great space for development.