Skin distortion is a long standing challenge in fingerprint matching, which causes false non-matches. Previous studies have shown that the recognition rate can be improved by estimating the distortion field from a distorted fingerprint and then rectifying it into a normal fingerprint. However, existing rectification methods are based on principal component representation of distortion fields, which is not accurate and are very sensitive to finger pose. In this paper, we propose a rectification method where a self-reference based network is utilized to directly estimate the dense distortion field of distorted fingerprint instead of its low dimensional representation. This method can output accurate distortion fields of distorted fingerprints with various finger poses. Considering the limited number and variety of distorted fingerprints in the existing public dataset, we collected more distorted fingerprints with diverse finger poses and distortion patterns as a new database. Experimental results demonstrate that our proposed method achieves the state-of-the-art rectification performance in terms of distortion field estimation and rectified fingerprint matching.
Estimating orientation field for latent fingerprints plays a crucial role in latent fingerprints recognition systems. Due to poor quality and small area of latent fingerprints, however, the performance of the state-of-the-art algorithms is still far from satisfactory. Considering the intrinsic characteristics of fingerprints that the distribution of orientation field varies with the fingerprint patterns, we propose an orientation field estimation algorithm for latent fingerprints based on residual learning using prior knowledge of fingerprint patterns. Specifically, statistical distribution models of orientation field, for different fingerprint patterns, are calculated based on a large database consisting of 14,000 fingerprints with good quality using clustering method. The residual orientation fields and reliability scores, indicating the consistency with different statistical orientation models, are estimated using a deep network, named RefNet. Then the final orientation field is obtained by fusing the estimations according to their corresponding reliability scores. Experimental results on the widely used latent database NIST SD27 demonstrate that the proposed algorithm provides higher orientation field estimation accuracy compared with the state-of-the-art methods, and by enhancing latent fingerprints using estimated orientation field, the identification performance is further improved.
The naked carp (Gymnocypris przewalskii) plays a central role in the ecosystem of the Qinghai Lake, the largest saline-alkaline lake in China. The adult naked carp migrates in large groups with high population density annually from the Qinghai Lake to the upstream freshwater rivers to spawn. Nevertheless, the responsiveness of the fish to local abiotic cues in the form of distribution patterns during migration across the riverine-lacustrine network of the Qinghai Lake region remains unknown. This knowledge gap has reduced efficiency in fish conservation and management efforts in the region. To address this issue, we carried out two field surveys from June to August, 2018, with the aid of unmanned aerial vehicles to a 200-m long back channel characterizing diverse morphological and hydraulic features on the migration route. Combined structure from motion photogrammetry and deep neural network techniques were used to establish a new workflow for detecting and extracting the profiles of fish individuals in large schools. The spatio-temporal distribution pattern of the fish demonstrated that the naked carp was attracted by hydraulic environments with high flow velocity or deep-water during migration. The diurnal variation of temperature and light could alter the preference for hydraulic environments of the fish. Our results highlight the crucial role of the interactions between river morphology and hydraulics, water temperature and light on the migration behaviours of the naked carp.
Abstract Background Patients with diabetes mellitus (DM) caused by obesity have increased in recent years. The impact of obesity on long-term outcomes in patients undergoing percutaneous coronary intervention (PCI) with or without DM remains unclear. Methods We retrospectively analysed data from 1918 patients who underwent PCI. Patients were categorized into four groups based on body mass index (BMI, normal weight: BMI < 25 kg/m 2 ; overweight and obese: BMI ≥ 25 kg/m 2 ) and DM status (presence or absence). The primary endpoint was the occurrence of major adverse cardiac and cerebrovascular events (MACCE; defined as all-cause death, myocardial infarction, stroke, and unplanned repeat revascularization). Results During a median follow-up of 7.0 years, no significant differences in MACCE, myocardial infarction, or stroke were observed among the four groups. Overweight and obese individuals exhibited lower all-cause mortality rates compared with normal-weight patients (without DM: hazard ratio [HR]: 0.54, 95% confidence interval [CI]: 0.37 to 0.78; with DM: HR: 0.57, 95% CI: 0.38 to 0.86). In non-diabetic patients, the overweight and obese group demonstrated a higher risk of unplanned repeat revascularization than the normal-weight group (HR:1.23, 95% CI:1.03 to 1.46). After multivariable adjustment, overweight and obesity were not significantly associated with MACCE, all-cause death, myocardial infarction, stroke, or unplanned repeat revascularization in patients with and without diabetes undergoing PCI. Conclusion Overweight and obesity did not demonstrate a significant protective effect on long-term outcomes in patients with and without diabetes undergoing PCI.
Compared to minutia-based fingerprint representations, fixed-length representations are attractive due to simple and efficient matching. However, fixed-length fingerprint representations are limited in accuracy when matching fingerprints with different visible areas, which can occur due to different finger poses or acquisition methods. To address this issue, we propose a localized deep representation of fingerprint, named LDRF. By focusing on the discriminative characteristics within local regions, LDRF provides a more robust and accurate fixed-length representation for fingerprints with variable visible areas. LDRF can be adapted to retain information within any valid area, making it highly flexible. The matching scores produced by LDRF also exhibit intuitive statistical characteristics, which led us to propose a matching score normalization technique to mitigate the uncertainty in the cases of very small overlapping area. With this new technique, we can maintain a high level of accuracy and reliability in our fingerprint matching, even as the size of the database grows rapidly. Our experimental results on 21 datasets containing over 140K fingerprints of various finger poses and impression types show that LDRF outperforms other fixed-length representations and is robust to sensing technologies and impression types. Besides, the proposed matching score normalization effectively reduces the false match rate (FMR) in large-scale identification experiments comprising over 5.11 million fingerprints. Specifically, this technique results in a reduction of two orders of magnitude compared to matching without matching score normalization and five orders of magnitude compared to prior works.
Touchscreens are the primary input devices for smartphones and tablets. Although widely used, the output of touchscreen controllers is still limited to the two-dimensional position of the contacting finger. Finger angle (or orientation) estimation from touchscreen images has been studied for enriching touch input. However, only pitch and yaw are usually estimated and estimation error is large. One main reason is that touchscreens provide very limited information of finger. With the development of under-screen fingerprint sensing technology, fingerprint images, which contain more information of finger compared with touchscreen images, can be captured when a finger touches the screen. In this paper, we constructed a dataset with fingerprint images and the corresponding ground truth values of finger angle. We contribute with a network architecture and training strategy that harness the strong dependencies among finger angle, finger region, finger type, and fingerprint ridge orientation to produce a top-performing model for finger angle estimation. The experimental results demonstrate the superiority of our method over previous state-of-the-art methods. The mean absolute errors of the three angles are 6.6 degrees for yaw, 7.1 degrees for pitch, and 9.1 degrees for roll, markedly smaller than previously reported errors. Extensive experiments were conducted to examine important factors including image resolution, image size, and finger type. Evaluations on a set of under-screen fingerprints were also performed to explore feasibility in real-world applications. Code and a subset of the data are publicly available.
Several studies have explored the estimation of finger pose/angle to enhance the expressiveness of touchscreens. However, the accuracy of previous algorithms is limited by large estimation errors, and the sequential output angles are unstable, making it difficult to meet the demands of practical applications. We believe the defect arises from improper rotation representation, the lack of time-series modeling, and the difficulty in accommodating individual differences among users. To address these issues, we conduct in-depth study of rotation representation for the 2D pose problem by minimizing the errors between representation space and original space. A deep learning model, TrackPose, using a self-attention mechanism is proposed for time-series modeling to improve accuracy and stability of finger pose. A registration application on a mobile phone is developed to collect touchscreen images of each new user without the use of optical tracking device. The combination of the three measures mentioned above has resulted in a 33% reduction in the angle estimation error, 47% for the yaw angle especially. Additionally, the instability of sequential estimations, measured by the proposed metric MAEΔ, is reduced by 62%. User study further confirms the effectiveness of our proposed algorithm.
Abstract. Bimodal runoff behavior, characterized by two distinct peaks in flow response, often leads to significant stormflow and associated flooding. Understanding and characterizing this phenomenon is crucial for effective flood forecasting. However, this runoff behavior has been understudied and poorly understood in semi-humid regions. In this study, we investigated the response characteristics and occurrence conditions of bimodal hydrograph based on the hydrometric and isotope data spanning 10 years in a semi-humid forested watershed in North China. The main findings include: 1) the onset of the bimodal hydrograph exhibits a threshold behavior, with delayed streamflow peaks occurring when the sum of event rainfall (P) and antecedent soil moisture index prior to the rainfall (ASI) exceeds 200 mm; 2) isotopic hydrograph separation reveals that delayed stormflow process is primarily driven by pre-event water, with increasing contributions of pre-event water during catchment wetting-up; 3) the dynamic variation in groundwater level precedes that of streamflow, establishing a hysteretic relationship wherein groundwater level peaks before streamflow during delayed stormflow. These findings, supported by onsite observations, emphasize the dominance of shallow groundwater flow in the generation of delayed stormflow.
DeepKG is an end-to-end deep learning-based workflow that helps researchers automatically mine valuable knowledge in biomedical literature. Users can utilize it to establish customized knowledge graphs in specified domains, thus facilitating in-depth understanding on disease mechanisms and applications on drug repurposing and clinical research. To improve the performance of DeepKG, a cascaded hybrid information extraction framework is developed for training model of 3-tuple extraction, and a novel AutoML-based knowledge representation algorithm (AutoTransX) is proposed for knowledge representation and inference. The system has been deployed in dozens of hospitals and extensive experiments strongly evidence the effectiveness. In the context of 144 900 COVID-19 scholarly full-text literature, DeepKG generates a high-quality knowledge graph with 7980 entities and 43 760 3-tuples, a candidate drug list, and relevant animal experimental studies are being carried out. To accelerate more studies, we make DeepKG publicly available and provide an online tool including the data of 3-tuples, potential drug list, question answering system, visualization platform.All the results are publicly available at the website (http://covidkg.ai/).Supplementary data are available at Bioinformatics online.
Latent fingerprint matching is a daunting task, primarily due to the poor quality of latent fingerprints. In this study, we propose a deep-learning based dense minutia descriptor (DMD) for latent fingerprint matching. A DMD is obtained by extracting the fingerprint patch aligned by its central minutia, capturing detailed minutia information and texture information. Our dense descriptor takes the form of a three-dimensional representation, with two dimensions associated with the original image plane and the other dimension representing the abstract features. Additionally, the extraction process outputs the fingerprint segmentation map, ensuring that the descriptor is only valid in the foreground region. The matching between two descriptors occurs in their overlapping regions, with a score normalization strategy to reduce the impact brought by the differences outside the valid area. Our descriptor achieves state-of-the-art performance on several latent fingerprint datasets. Overall, our DMD is more representative and interpretable compared to previous methods.
'/%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)