Finger Orientation Research Articles

3D Finger Rotation Estimation from Fingerprint Images

Various touch-based interaction techniques have been developed to make interactions on mobile devices more effective, efficient, and intuitive. Finger orientation, especially, has attracted a lot of attentions since it intuitively brings three additional degrees of freedom (DOF) compared with two-dimensional (2D) touching points. The mapping of finger orientation can be classified as being either absolute or relative, suitable for different interaction applications. However, only absolute orientation has been explored in prior works. The relative angles can be calculated based on two estimated absolute orientations, although, a higher accuracy is expected by predicting relative rotation from input images directly. Consequently, in this paper, we propose to estimate complete 3D relative finger angles based on two fingerprint images, which incorporate more information with a higher image resolution than capacitive images. For algorithm training and evaluation, we constructed a dataset consisting of fingerprint images and their corresponding ground truth 3D relative finger rotation angles. Experimental results on this dataset revealed that our method outperforms previous approaches with absolute finger angle models. Further, extensive experiments were conducted to explore the impact of image resolutions, finger types, and rotation ranges on performance. A user study was also conducted to examine the efficiency and precision using 3D relative finger orientation in 3D object rotation task.

Just visual context or part of the gesture? The role of arm orientation in bent pointing interpretation

Pointing gestures can take on different shapes. For example, people often point with a bent wrist at a referent that is occluded by another object. We hypothesized that while the extrapolation of the index finger is the most important visual cue in such bent pointing gestures, arm orientation is affecting interpretations as well. We tested two competing hypotheses. First, the arm could be processed as a less reliable but additional direction cue also indicating the referent. Consequently, the index finger extrapolation would be biased towards the arm direction (assimilation effect). Second, the arm could be perceived as visual context of the index finger, leading to an interpretation that is repulsed from the arm direction (contrast effect). To differentiate between both, we conducted two experiments in which arm and finger orientation of a virtual pointer were independently manipulated. Participants were asked to determine the pointed-at location. As expected, participants based their interpretations on the extrapolation of the index finger. In line with the second hypothesis, the more the arm was oriented upwards, the lower the point was interpreted and vice versa. Thus, interpretation pattern indicated a contrast effect. Unexpectedly, gestures with aligned arm and index finger deviated from the general contrast effect and were interpreted linearly compared to bent gestures. In sum, the experiments show that interpretations of bent pointing gestures are not only based on the direction of the index finger but also depend on the arm orientation and its relationship to the index finger orientation.

Examining constraints on embodiment using the Anne Boleyn illusion.

Using a mirror box, the concurrent stroking of the lateral side of the fifth finger behind the mirror along with stroking the empty space next to the mirror-reflected hand's fifth finger results in a strong sense of having a sixth finger-the Anne Boleyn illusion. We used this illusion to understand what constraints illusory embodiment. In Experiment 1, we manipulated the anatomical constraints, posture, and stroking of the sixth finger, along with other variants. Given evidence from other body illusions, we predicted no illusory embodiment in conditions in which the sixth finger was created in a manner incompatible with a typical hand, when the mirror and viewed hands were in different posture, and when stroking differed. Surprisingly, the illusion was persistent in most variants, including those with curved fingers, elongated fingers, and even with mismatches between the posture of the viewed and hidden hand. In Experiment 2, we manipulated the orientation, shape, and length of the illusory sixth finger, presenting more extreme versions of the illusion. The illusion was significantly diminished only when the sixth finger was far from the hand, or in a very implausible posture. This evidence suggests that body representations are extremely flexible and allow for embodiment of empty space in conditions not seen in other body illusions. We suggest that bottom-up information from concurrent visuotactile input, combined with reduced constraints provided by the "blank canvas" of empty space, results in a particularly robust illusion. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Patient Assistance System Based on Hand Gesture Recognition

We propose a two-stage hand gesture recognition architecture to support a patient assistance system. Some medical conditions limit mobility, and the patient must rely on medical staff to meet their needs. In such cases, a phone or intercom is not convenient to call for help. A vision-based system operated by changing the orientation of fingers can be used to send specific messages without making arm movements. However, vision-based hand gesture recognition is hindered by occlusion, background clutter, and variations in illumination. Therefore, we developed a two-stage architecture: The first stage produces a saliency map to simplify recognition and the second stage performs classification. A novel combined loss function optimizes the saliency detection model and makes the saliency map more precise. An adaptive kernel-based channel attention layer is used to emphasize salient features. The proposed architecture achieved precise saliency detection on four benchmark datasets and high-accuracy recognition on two. We designed an interface for patients to send specific messages to the medical staff using hand gestures. The interface help patients request assistance and connect with medical staff without leaving the bed or involving a third party.

Robotic Fingers in Reach-to-Grasp Tasks of Rehabilitation

The REHAROB robotic upper limb rehabilitation system was improved with a custom-designed and developed hand/finger therapy module. The new module extends the scope of the applicable motion therapy from passive to active reach-to-grasp activities of daily living tasks, and the range of treated anatomical joints was also extended to every proximal and distal upper limb anatomical joint. Finger exercising and object grasping are supported with a pair of two degree-of-freedom (DOF) robotic fingers. One of the robotic fingers moves the index/middle/ring fingers together, whereas the other robotic finger moves the thumb. A novel hypothesis was established, analyzed, and tested for setting the orientation of the robotic finger moving the thumb. The robotic thumb is not aligned with the patient's thumb; its orientation is optimized in the patient's hand reference system to maximize the efficiency in the opposite grasping task. While most concurrent systems utilize virtual objects for grasping tasks, the REHAROB system exercises five carefully selected reach-and-grasp type activities of daily living (ADL) with real objects. Actuating the human finger phalanges through custom development finger orthoses is described. An advanced feature of the hand/finger therapy module is the left-right hand side changeover by only alternating the orientation of the robotic fingers and exchanging the finger orthoses.

HCI Based Virtual Controlling System

Abstract: Researchers around the globe are working on making our devices more interactive and making them function with minimal physical contact in this research project. The proposed system is an interactive computer system that can operate without a physical keyboard or mouse. This system will benefit everyone, particularly immobilized people with special needs operating a physical keyboard and mouse. So in the system, they have developed an interface that uses visual hand-gesture analysis. These gestures are used to assist those who are having trouble controlling or operating computers or gadgets. The model is being developed in such a way that it can assist us in recognizing and implementing it. Regarding hand gestures, our interface uses OpenCV, Python, and computer vision algorithms that can detect different finger orientations, distinguish the user's hand from the background, and distinguish significant hand movements from unwanted hand movements. Keywords: Machine Learning, Computer Vision, Image Processing, OpenCV, Human-Computer Interaction (HCI)

Step-by-step optimisation of robotic-assisted radical prostatectomy using augmented reality

ABSTRACTIntroduction:Surgical training will be complemented by digitalisation, as the COVID 19 pandemic continues (1). Proximie is an augmented reality (AR) platform that can display up to 4 native camera views, with live or semi live telementoring. It can optimise ergonomics of the surgeon at the console (2), and robotic instrument orientation. We describe the utilisation of Proximie as a step-by-step guide in a robotic assisted radical prostatectomy (RARP).Surgical Technique:Author V. P. performed a transperitoneal multiport da Vinci Xi RARP with the Proximie platform: a laptop computer, multiple HD webcams, microphones and speakers. Using an HDMI cable to the Intuitive Surgical tower, output display from the console and an additional laparoscopic tower is shown. Each webcam was mounted to the side armrests of the console, directed at the surgeon's hands. An independent ‘drop in’ laparoscope via an additional 5mm left upper quadrant port was utilised. Observers can visualise the AR platform's recordings on a laptop and/or smartphone. A PTZ (pan-tilt-zoom) camera can capture the operating room, bedside assistant, ports and patient position. Our video demonstrates three of four camera views for posture, forearm, wrist, hand, and finger orientation, relative to the translated robotic steps. A pincer grasp of the endowrist manipulator during anastomosis allows optimal robotic wrist rotation. The second laparoscopic camera view demonstrated intracorporeal angles of robotic arm and bedside assistant's instrument position for critical steps such as nerve sparing and anastomosis (3). The console time was 100 minutes, no intraoperative complications, or delay in image transmission occurred with utilising the platform.Considerations:An AR platform can create deeper learning for RARP in real time or recorded sessions. Two-way verbal and visual communication with ability to annotate on screen, allows long distance mentoring. The platform's utility can be accessed in anywhere, to project surgeons beyond their immediate environment. This allows for democratisation of access to high volume institutions and their evolution of techniques (4), to assist patients globally. Potential developments are artificial intelligence (AI) networks analysing repository of such recorded data, to identify intraoperative hand motion and robotic instrument tracking. AR is a pertinent building block to enhance robotic training, skill dissemination, precision medicine (5) and surgery overall.

HOS-FingerCode: Bispectral invariants based contactless multi-finger recognition system using ridge orientation and feature fusion

Fingerprints are a commonly used biometric for civil and forensic applications and the majority of them are captured by contact sensors. More recently, there is a trend toward the use of high resolution digital and video cameras to acquire contactless images. However, the processing of such images is challenging because of intra-class variations such as pose, rotation, translation and scaling due to hand orientation and deformation. They can, however, facilitate the use of full finger or multiple fingers or hand information rather than the fingertip alone. Although some recent research has investigated representation of full finger and hand information under relaxed conditions, current state-of-the-art methods are unable to address transformation issues without manual intervention. To overcome the issues of relaxed constrained nature, geometrical variations and invariant feature encoding regardless of finger orientation, we have used higher order spectral features (HOS-FingerCode) from ridge patterns. We employ an automated method to detect key points and constructs a graph of key-lines. Ridge orientation profiles along with the selected key-lines provide a 1D signal from which bispectral invariant features are extracted. In addition, two fusion schemes are proposed where fusion from multiple key lines combinations and feature fusion from multiple key line configurations are investigated to extract ridge pattern profile for a better performance. The features are pooled from multiple fingers by concatenation or compact bilinear pooling to yield the HOS-FingerCode for multi-finger biometrics. For a single-finger system using 3369 high resolution index and middle finger images, the algorithm better performs when different key lines configurations are fused. For geometrical variations of the images, the system is most impacted by pose than rotation and scale changes of the images, and is tolerant to such variations. For the multi-finger system using 3111 images from both fingers, the algorithm has achieved 93.8% TPR and 97.95% classification accuracy at a setting of 2% FPR. For the feature fusion scheme proposed for multi-finger system using concatenation and compact bilinear pooling, concatenation of features performs better than the compact bilinear pooling. Since multi-finger recognition system performed better than the single finger biometrics system, it can help to diminish the high traffic scenario in busy premises and will facilitate the soft identity verification.

A Low-Cost Wearable Hand Gesture Detecting System Based on IMU and Convolutional Neural Network.

In this paper, a low-cost wearable hand gesture detecting system based on distributed multi-node inertial measurement units (IMUs) and central node microcontroller is presented. It can obtain hand kinematic information and transmit data to the remote processing terminal wirelessly. To have a comprehensive understanding of hand kinematics, a convolutional neural network (CNN) model on the terminal is proposed to recognize and classify gestures and the modified Denavit-Hartenberg notation is used to acquire finger spatial locations. The experiment has not only completed a variety of gesture recognitions, but also captured and displayed the orientation and posture of a single finger. The prototype can be used in various occasions such as hand rehabilitation evaluation and human-computer interaction.

Structure of variability in scanning movement predicts braille reading performance in children

Among children learning to read braille, we asked whether the quantitative kinematics of scanning movements of the reading finger would be related to the proficiency of braille reading. Over a period of 12 months, we recorded the position and orientation of the reading fingers of eight congenitally or early blind children. We found that the strength of long-range power-law temporal correlations in the velocity fluctuations increased with performance in braille reading. In addition, we found that the variability of the angular orientation of the reading finger that affects the contact region on the fingerpad was negatively related to braille reading performance. These results confirm that the quantitative kinematics of finger scanning movements were related to functional performance in braille reading. The results add to the growing body of evidence that long-range temporal correlations in exploratory behavior can predict perceptual performance, and that scanning movements that center important tactile information on the small, high resolution area contribute to the pickup of information.

Sign Language Recognition System

In this paper an attempt has been made to design a sign language recognition system. An intelligent glove has been designed to automate the communication between a deaf-mute with others by converting sign language into speech or understandable language. The sensory gloves provide data of the human hand shape or movement and translate it to text and speech. It comprises hardware and software for translating sensor data. It is wearable devices that can be put on human hands and convert hand's gestures into signs letter by letter and send the data into the firebase for further processing. The glove is equipped with flex sensors and an inertial measurement unit to recognize the movement by monitoring the finger orientation and hand motion in three-dimensional spaces that senses a person's gestures in the form of finger bend and hand fist tilt. The Hall sensor has been used to process and collect data for training and model development. The three different machine learning algorithms, i.e., support vector machine, Naive Bayes, decision tree, have been used for analysis. It has been observed that the support vector machine has the highest accuracy, i.e., 90%. After Analyzing, the data has been sent to the speech converting function, and then audible results have been produced.

Molecular phylogeny and species delimitation of the freshwater prawn Macrobrachium pilimanus species group, with descriptions of three new species from Thailand.

Specific status and species boundaries of several freshwater prawns in the Macrobrachium pilimanus species group remain ambiguous, despite the taxonomic re-description of type materials and additional specimens collected to expand the boundaries of some species. In this study, the “pilimanus” species group of Macrobrachium sensu Johnson (1958) was studied using specimens collected from montane streams of Thailand. Molecular phylogenetic analyses based on sequences of three molecular markers (COI, 16S and 18S rRNA) were performed. The phylogenetic results agreed with morphological identifications, and indicated the presence of at least nine putative taxa. Of these, six morphospecies were recognised as M. malayanum, M. forcipatum, M. dienbienphuense, M. hirsutimanus, M. eriocheirum, and M. sirindhorn. Furthermore, three morphologically and genetically distinct linages were detected, and are described herein as M. naiyanetri Siriwut sp. nov., M. palmopilosum Siriwut sp. nov. and M. puberimanus Siriwut sp. nov. The taxonomic comparison indicated wide morphological variation in several species and suggested additional diagnostic characters that are suitable for use in species diagnoses, such as the shape and orientation of fingers, the rostrum form, and the presence or absence of velvet pubescence hairs and tuberculated spinulation on each telopodite of the second pereiopods. The “pilimanus” species group was portrayed as non-monophyletic in both ML and BI analyses. The genetic structure of different geographical populations in Thailand was detected in some widespread species. The species delimitation based on the four delimitation methods (BIN, ABGD, PTP and GMYC) suggested high genetic diversity of the “pilimanus” species group and placed the candidate members much higher than in previous designations based on traditional morphology. This finding suggests that further investigation of morphological and genetic diversity of Southeast Asian freshwater prawns in the genus Macrobrachium is still required to provide a comprehensive species list to guide efforts in conservation and resource management.

BLT Gripper: An Adaptive Gripper With Active Transition Capability Between Precise Pinch and Compliant Grasp

Achieving both the precise pinching and compliant grasping capabilities is a challenging goal for most robotic hands and grippers. Moreover, an active transition between the pinching pose and the grasping pose is difficult for limited degrees-of-freedom (DOF) hands or grippers. Even when using high DOF robotic hands, it requires a substantial amount of control effort and information from tactile sensors. In this letter, a 3-finger, 5-DOF adaptive gripper with active transition capability is presented. Each finger is composed of a minimum number of components using one rigid link, one belt, one fingertip frame and one motor for flexion motion. This simple and unique finger structure enables precise parallel pinching and highly compliant stable grasping with evenly distributed pressure. The other two motors are used for fingertip angle adjustment and change of the finger orientation respectively. Thorough kinematic and force analysis with detailed descriptions of the mechanical design clearly shows controllable transition property and stable grasping performance. The experiments including the grasping force and pressure measurement verify the performance of the proposed gripper and prove the practical usefulness for real-world applications.

Grassmann manifold based dynamic hand gesture recognition using depth data

In this paper, we propose a novel Grassmann manifold based framework for dynamic hand gesture recognition from depth data. Automated dynamic hand gesture recognition is important for improving man-machine communication and understanding human behavior. It finds various applications such as human computer interaction, ambient assisted living, automated driver assisted systems. We use depth data or skeleton information to detect the fingertip and store the fingertip points to create the trajectory. In fingertip detection using depth data first we detect the hand using the depth data and used hand shape properties such as finger thickness, finger length, finger width and finger orientation angle to find the shape of the hand. If skeleton data is available we use skeleton information to detect the fingertip in each frame. Then geometrical features are extracted and a unique gesture subspaces created using SVD for each feature vector matrix of each gesture set. These gesture subspaces lie on a Grassmann manifold and capture the intra-class variations and increase the inter-class discriminatory power. We apply Grassmann manifold based discriminant analysis for recognizing each test gesture. We perform experiments on standard datasets and the results show that we have achieved recognition accuracy comparable to the state-of-the-art.

A Möbius transformation based model for fingerprint minutiae variations

When an individual’s fingerprint is scanned, although the global fingerprint pattern is unchanged, at the local level, between different scans the minutiae pattern may vary. Minutiae translation and rotation are caused by changing finger orientation and position shift during fingerprint acquisition. Minutiae patterns may also suffer non-linear distortion due to finger skin elasticity. Despite a variety of approaches to detecting deformations in fingerprint images, there has been no method available for capturing minutiae variations between two impressions of the same finger in a unified model. In this paper we address this issue by proposing a unified model to represent minutiae variations between fingerprint scans and formulate the changes to minutiae feature patterns. We identify the Möbius transformation as a good candidate for modelling minutiae translation, rotation and non-linear distortion, that is, different types of minutiae variations are described in a single model. Not only do we mathematically prove that the Möbius transformation based model is a unified model for capturing minutiae variations, but we also experimentally verify the effectiveness of this model using a public database.

Structure, emplacement mechanism and magma-flow significance of igneous fingers – Implications for sill emplacement in sedimentary basins

Field and seismic observations show that numerous sills exhibit lobate morphologies. Each lobe corresponds to a distinct igneous segment exhibiting a finger-like shape, the long axis of which is commonly interpreted as a magma-flow indicator. Robust understanding of the emplacement mechanisms of finger-shaped sills, and direct observations supporting finger orientation as magma-flow indicator are lacking. In this paper, we present the results of detailed structural mapping on an exceptional, easily accessible 1-km long outcrop in the Neuquén Basin, Argentina, that exhibits a sill, its contacts and the structures in the finely layered sedimentary host rock. We show that the sill is made of distinct segments that grew, inflated or coalesced. We also demonstrate that the fingers were emplaced according to the viscoelastic fingering or viscous indenter models, with no field evidence of tensile elastic fracture mechanism as commonly assumed in mechanical models of sill emplacement. We identified new structural criteria at the intrusion's contacts for inferring magma flow direction during the magma emplacement. Our small-scale structural observations carried out on a seismic-scale outcrop have the potential to considerably aid the structural interpretation of seismic data imaging igneous sills, i.e. to fill the standard gap between outcrop-scale field observations and seismic-scale geophysical data.

ZFAS1: A novel vital oncogenic lncRNA in multiple human cancers.

Long noncoding RNAs (lncRNAs) are a class of noncoding, endogenous, single‐stranded RNAs longer than 200 nucleotides in length that are transcribed by RNA polymerase II. Mounting evidence has indicated that lncRNAs play key roles in several physiological and pathological processes by modifying gene expression at the transcriptional, posttranscriptional, epigenetic, and translation levels. Many reports have demonstrated that lncRNAs function as potential oncogene or tumour suppressors and thus play vital regulatory roles in tumourigenesis and tumour progression. ZNFX1 antisense RNA 1 (ZFAS1), a novel lncRNA transcribed in the antisense orientation of zinc finger NFX1‐type containing 1(ZNFX1), was found to be increased in multiple cancers, such as gastric cancer and hepatocellular carcinoma, contributing to cancer development and progression. In the present review, we summarized recent progression on study of the functions and underlying molecular mechanisms of ZFAS1 related to occurrence and development of multiple cancers.

A Wearable Three-Axis Tactile Sensor for Human Fingertips

Human training data can scaffold robot imitation learning. However, most datagloves only record joint angles, but for many tasks, force control is more important. They also cover the human skin, thereby making a natural interaction of the human with the object impossible. This letter suggests a wearable sensor that can measure the three-axis force vector exerted at the fingertip, without covering the skin that contacts the object. Using two small-sized Hall effect based three-axis sensors mounted on the sides of the fingertips, the finger pad deformation resulting from the force vector acting on the fingertip is measured. Experiments with ten subjects show that the force vectors can be measured with reasonable precision. Furthermore, the influence of the Earth's magnetic field and the finger's orientation on the finger tactile sensor's measurement has been reduced from $\pm 4N$ to $\pm 1N$ .

Sign Language Interpreter

The world’s population now stands at 7.6 billion. Out of these, millions of people are not blessed with the ability to hear or/and speak. Although speech becomes a barrier between these people and other abled persons but, there are three types of sign languages which help to decrease this gap. There are about 250 types of sign language interpreters which translate their language for other people. This paper describes a similar type of device aimed to facilitate a bridge between the two groups of population. It develops a wearable device having motion detecting sensors to sense the orientation and direction of fingers in order to recognize the gesture made by the user. Also, it transmits the gesture to a wireless display unit through a Bluetooth module which displays the corresponding word to the gesture made on an LCD. Thus, in this way, sign-to-text conversion is carried out. The heart of the device is Atmega328 with flash memory of 32k bytes enabling us to store a wide range of words. The designed system was capable to recognize the pre-stored gestures according to ASL (American Sign Language). Prototype was designed for one hand and it was tested on a few gestures.

Structural performance evaluation of finger-jointed rubberwood manufactured by factories in Thailand

In order to utilize finger-jointed rubberwood as raw material for the production of structural wood products, the finger joint efficiencies of rubberwood specimens taken from three factories in Thailand were evaluated. This study investigated the finger profile, modulus of rupture (MOR), and modulus of elasticity (MOE) of finger-jointed rubberwood from all selected factories. The effect of finger orientations (vertical and horizontal) on MOR and MOE values was also examined. The results showed that all selected factories used the same finger profile for manufacturing finger jointing of rubberwood samples. The finger orientations had no noticeable effect on the MOR and MOE values. The MOR values of finger-jointed rubberwood obtained from all selected factories were different. They ranged from 55 to 78 MPa. A primary cause of failure for specimens with lower MOR values was the poor surface bonding of fingers. The MOE values of samples were similar for all selected factories ranging from 9,710 to 12,200 MPa. According to BS EN 338 (2016), finger jointed rubberwood from some factories was inappropriate for production of high strength structural wood products.