Standard Color Camera Research Articles

Optical High-Speed Rolling Mark Detection Using Object Detection and Levenshtein Distance

This paper presents an automated high-speed rolling mark recognition system for railroad rails utilizing image processing techniques. Rolling marks, which consist of numbers, letters, and special characters, were engraved into the rail web as 3D information. These rolling marks provide crucial details regarding the rail manufacturer, steel quality, year of production, and rail profile. As a result, they empower rail infrastructure managers to gain valuable insights into their infrastructure. The rolling marks were captured using a standard color camera under dark field illumination. The recognition of individual numbers, letters, and special characters was achieved through state-of-the-art deep neural network object detection, specifically employing the YOLO architecture. By leveraging reference rolling marks, the detected characters can then be accurately interpreted and corrected. This correction process involves calculating a weighted Levenshtein distance, ensuring that the system can identify and rectify partially misidentified rolling marks. Through the proposed system, the accurate and reliable identification of rolling marks was achieved, even in cases in which there were partial errors in the detection process. This novel system thus has the potential to substantially improve the management and maintenance of railroad infrastructure.

Improved Cherenkov Imaging across a Wide Range of Skin Pigmentation Levels for the Inclusion of Diverse Patient Populations

<h3>Purpose/Objective(s)</h3> Cherenkov imaging has recently emerged as a tool to visualize radiation dose and detect treatment incidents in real-time. The principle relies on the proportionality between Cherenkov light emission and dose deposited within tissue, introducing an additional opportunity to use the image information as a 2D surface dose map. The Cherenkov-dose relationship is affected by tissue optical properties, and recent studies have utilized routine CT-scans to correct for optical attenuation due to fibroglandular and adipose tissues, and blood volume, in the breast. However, the bulk of the reported <i>in vivo</i> data is limited to light-skinned patients due to limitations of geographical demographics. It is well documented that melanin is an optical absorber of Cherenkov photons, yet no prior <i>in vivo</i> work has been done to characterize the effect of different skin tones on Cherenkov dosimetry. We hypothesize that skin color information can be used to correct Cherenkov images for optical attenuation relating to melanin absorption across a diverse patient population. <h3>Materials/Methods</h3> A multi-institutional partnership was developed aiming to collect data from a more diverse patient population. A time-gated, intensified CMOS camera installed in the ceiling of a treatment room collects Cherenkov images of breast RT patients during their treatment. A standard color camera module is mounted adjacent to the Cherenkov camera to collect a color background image. Following a process developed with phantom studies, the two images are co-registered together in post-processing, RGB values are extracted from the skin over the treatment regions to calculate the relative skin luminance, L, and this value is compared to the optical emission from the corresponding Cherenkov image. A linear regression model is then used to assess the proportionality between L and Cherenkov emission intensity. <h3>Results</h3> The Cherenkov imaging system is being deployed in a clinic with a catchment area demographic of 67.4% Non-Hispanic White, an increase in diversity from 89.4% at the home institution. A clear relationship exists between skin color information and Cherenkov intensity, which cannot be determined like the bulk tissue properties from the CT scan alone. We found that L, derived from a wide range of melanin concentrations, is linearly proportional to Cherenkov intensity (R² = 0.98). Each Cherenkov image can be corrected based on skin luminance to improve the Cherenkov-dose relationship across various skin types. <h3>Conclusion</h3> The effect of both inter- and intra-patient skin color variations on the Cherenkov intensity can be corrected by using a linear regression model to increase the linearity between Cherenkov and deposited dose. This work will be used to improve our current CT-correction algorithm for all skin types and presents a major step towards expanding the inclusivity of Cherenkov imaging as a dosimetric tool for every patient.

Action Recognition in Physical Rehabilitation - A review

Recognizing actions from film clips or images is a difficult procedure due to difficulties such as crowded backgrounds, occlusions, differences in scale, viewpoint, colour, and look.Various technologies, such as surveillance cameras, human - machine interface, and robotics for human activity description, requires a simultaneous action recognition system. A standard colour camera can identify a moving silhouette in the foreground, but precise assessment of human movements is difficult due to lack of complete three dimensional knowledge of the scene(1). Cameras like RGB-D can combine a depth map based on a colour image and the reflection of an infra red sensor light; they can help in solving the problem. The main aim in rehabilitation is correctness of a movement may be certified without being exceedingly accurate. The precise of skeleton estimation wasalso tested in order to better comprehend the potential clinical applications. The motion detection is a promising area for advancement in fields such as physiotherapy(2).

Low-cost camera-based continuous-wave laser detection

A low-cost camera-based method of detecting continuous-wave (cw) lasers has been developed at Defence Science and Technology Laboratory. The detector uses a simple optical modification to a standard color camera combined with image processing techniques to distinguish lasers from other illumination sources and measure the wavelength, direction, and irradiance of the laser light. Such a detector has applications in collecting information on aircraft laser dazzle incidents, providing the evidence required to inform on aircrew laser exposure events and to assess if engagements are eye safe. A prototype has been developed using entirely commercially available off-the-shelf (COTS) components, costing ≈£600, and assessed in laboratory conditions with the capability of measuring laser wavelengths to ±5 nm and irradiances to ±10 % . A realistic handheld laser engagement scenario, using a range of relevant wavelengths and irradiances, was simulated during the Moonraker trial where the prototype was capable of measuring laser wavelengths to an accuracy of ±10 nm and peak irradiances to ±25 % . All laser engagements were detected over a total data collection period of 9 h with zero false alarms. Comparisons were made with a COTS laser detector, which showed an equivalent performance. This technology offers a low-cost approach to cw laser detection, which is capable of extracting a range of parameters while maintaining a relatively wide field of view and angular resolution.

Non-Contact Pulse Detector using Video Analytics

Non-contact pulse detector used for heart beat measurement based on computer vision, where a standard color camera captures the plethysmographic signal and the heart rates are processed and estimated dynamically. It is important that the quantities are taken in a non-invasive manner, which is invisible to the patient. Presently, many methods have been proposed for non-contact measurement. The proposed method based on the computer vision technique is enhanced to overcome the above drawbacks and it requires low computational cost. Many of the hospitals are using surveillance cameras, from these cameras we can monitor the video of the patients waiting in the queue. The camera is attached in the patients’ waiting room and the faces of the patients are monitored. Many factors are considered in the phases of image acquisition, as well as in the plethysmographic signal development, pre-processing and filtering. The pre-filter step uses numerical analysis techniques to cut the signal offset. The proposed method decouples the heart rate from the plethysmographic signal frequency. The proposed system helps in detecting the heart rate of a Patient who is waiting in queue for longer time. Based on the heart rate the seriousness of patient is identified and giving the preference to the patient and treatment will be started, with this the patient will be in safe side.

ImageBreed: Open‐access plant breeding web–database for image‐based phenotyping

AbstractHigh‐throughput image‐phenotyping promises to accelerate the rate of genetic improvement in plant breeding through varietal selections informed by longitudinal growth models. To facilitate routine analyses and to drive breeding decisions, data integration is critical for effective management of germplasm, field experiment design, phenotyping, tissue sampling, genotyping, aerial‐phenotyping campaigns, image files, and geo‐spatial information. To this end, ImageBreed provides a software solution for end‐to‐end image‐based phenotyping integrated into the Breedbase plant breeding system. ImageBreed provides open‐source orthophotomosaic construction for raw image captures from standard color cameras and from the MicaSense RedEdge multispectral camera. Additionally, previously assembled orthophotomosaic raster images can be uploaded. Orthophotomosaic images allow for streamlined extraction of plot‐polygon images; however, ImageBreed plot‐polygon images can also be extracted directly from raw aerial image captures. A web–database interface streamlines assignment of plot‐polygon images from the orthophotomosaic or raw aerial‐captures to the field experiment design. Image processes spanning Fourier‐transform filtering, thresholding, and vegetation index masking are applied to reduce noise in extracted phenotypes. Summary‐statistic phenotypic values are extracted for every observed plot‐polygon image using a structured ontology. Plot‐polygon images are queryable against genotypic, phenotypic, and experimental design information for training of machine learning models and for driving breeding decisions in varietal advancement. ImageBreed is publicly available at http://imagebreed.org and built on the open‐source Breedbase system (https://github.com/solgenomics/sgn); all image‐processing scripts are available at https://github.com/solgenomics/DroneImageScripts and via a Docker image. All data deposited in http://imagebreed.org are publicly available for longitudinal model training and for driving future breeding decisions.

Object segmentation based on refractive index estimated by polarization of specular reflection

Polarization is steadily attracting attention in machine vision due to its ability to capture the information not readily available in standard color or greyscale camera. This paper aims to segmentation object by means of approximate value of refractive index. First, estimate the approximate refractive index by using relationships between the degree of polarization for specular reflection and zenith angle and refractive index, and then the approximate refractive index data are clustered to achieve object segmentation. The experimental results show that the segmentation method has great speed and accuracy in calculation.

Non-Contact, Simple Neonatal Monitoring by Photoplethysmography.

This paper presents non-contact vital sign monitoring in neonates, based on image processing, where a standard color camera captures the plethysmographic signal and the heart and breathing rates are processed and estimated online. It is important that the measurements are taken in a non-invasive manner, which is imperceptible to the patient. Currently, many methods have been proposed for non-contact measurement. However, to the best of the authors’ knowledge, it has not been possible to identify methods with low computational costs and a high tolerance to artifacts. With the aim of improving contactless measurement results, the proposed method based on the computer vision technique is enhanced to overcome the mentioned drawbacks. The camera is attached to an incubator in the Neonatal Intensive Care Unit and a single area in the neonate’s diaphragm is monitored. Several factors are considered in the stages of image acquisition, as well as in the plethysmographic signal formation, pre-filtering and filtering. The pre-filter step uses numerical analysis techniques to reduce the signal offset. The proposed method decouples the breath rate from the frequency of sinus arrhythmia. This separation makes it possible to analyze independently any cardiac and respiratory dysrhythmias. Nine newborns were monitored with our proposed method. A Bland-Altman analysis of the data shows a close correlation of the heart rates measured with the two approaches (correlation coefficient of 0.94 for heart rate (HR) and 0.86 for breath rate (BR)) with an uncertainty of 4.2 bpm for HR and 4.9 for BR (k = 1). The comparison of our method and another non-contact method considered as a standard independent component analysis (ICA) showed lower central processing unit (CPU) usage for our method (75% less CPU usage).

Saphir

The goal of optical playback of analogue audio discs records has been pursued since at least 1929. Several different approaches have been demonstrated to work. But in most cases the playback quality is worse than using mechanical playback. The Saphir process uses a specifically designed colour illuminator that exploits the reflective properties of the disc material to highlight subtle changes in orientation of the groove walls, even at highest frequencies (20kHz). A standard colour camera is used to collect rings of pictures from the disc. Audio signal is extracted from the collected pictures automatically, under user control. When colour signal is not useable, track slope can be used as an alternative. The process is slow—several hours per disc—but has a wide range of operation on recorded and printed discs, from earliest Berliner recordings to recent vinyl records, and its strength is at decoding direct-recording lacquer discs. An Elementary Shortest Path Solver with a reward (negative cost) on the number of turns is used to re-connect all the sub-tracks obtained, allowing to reconstruct, with limited human intervention, the correct playback order. We describe the approach and present the main advantages and drawbacks. The process was used to play back a number of extremely damaged (broken, de-laminated…), physically unplayable records.

Plant Phenotyping: An Active Vision Cell for Three-Dimensional Plant Shoot Reconstruction.

Three-dimensional (3D) computer-generated models of plants are urgently needed to support both phenotyping and simulation-based studies such as photosynthesis modeling. However, the construction of accurate 3D plant models is challenging, as plants are complex objects with an intricate leaf structure, often consisting of thin and highly reflective surfaces that vary in shape and size, forming dense, complex, crowded scenes. We address these issues within an image-based method by taking an active vision approach, one that investigates the scene to intelligently capture images, to image acquisition. Rather than use the same camera positions for all plants, our technique is to acquire the images needed to reconstruct the target plant, tuning camera placement to match the plant's individual structure. Our method also combines volumetric- and surface-based reconstruction methods and determines the necessary images based on the analysis of voxel clusters. We describe a fully automatic plant modeling/phenotyping cell (or module) comprising a six-axis robot and a high-precision turntable. By using a standard color camera, we overcome the difficulties associated with laser-based plant reconstruction methods. The 3D models produced are compared with those obtained from fixed cameras and evaluated by comparison with data obtained by x-ray microcomputed tomography across different plant structures. Our results show that our method is successful in improving the accuracy and quality of data obtained from a variety of plant types.

Multispectral analysis tools can increase utility of RGB color images in histology

Multispectral imaging (MSI) is increasingly finding application in the study and characterization of biological specimens. However, the methods typically used come with challenges on both the acquisition and the analysis front. MSI can be slow and photon-inefficient, leading to long imaging times and possible phototoxicity and photobleaching. The resulting datasets can be large and complex, prompting the development of a number of mathematical approaches for segmentation and signal unmixing. We show that under certain circumstances, just three spectral channels provided by standard color cameras, coupled with multispectral analysis tools, including a more recent spectral phasor approach, can efficiently provide useful insights. These findings are supported with a mathematical model relating spectral bandwidth and spectral channel number to achievable spectral accuracy. The utility of 3-band RGB and MSI analysis tools are demonstrated on images acquired using brightfield and fluorescence techniques, as well as a novel microscopy approach employing UV-surface excitation. Supervised linear unmixing, automated non-negative matrix factorization and phasor analysis tools all provide useful results, with phasors generating particularly helpful spectral display plots for sample exploration.

Measuring Heart and Breath Rates by Image Photoplethysmography using Wavelets Technique

This paper presents a method for contactless measuring of heart and respiratory rates by Photoplethysmography image technique through wavelets. This method is imperceptible to the humans, with the additional advantages of a low computational cost and high tolerance to artifacts. The approach uses the analysis of the plethysmographic signal collected by a standard color camera to estimate the heart and breathing rates. This system uses a pre-filter to reduce the offset of the averaged signal through numerical analysis techniques and the signal is processed by wavelets. Additionally, the measurement of respiration rate is decoupled from the sinus arrhythmia. A good estimate of the vital signs on ten individuals was obtained. The Bland-Altman analysis shows a strong correlation between measurements by standard systems and our approach.

Monitoring plant condition and phenology using infrared sensitive consumer grade digital cameras

Consumer-grade digital cameras are recognized as a cost-effective method of monitoring plant health and phenology. The capacity to use these cameras to produce time series information contributes to a better understanding of relationships between environmental conditions, vegetation health, and productivity. In this study we evaluate the use of consumer grade digital cameras modified to capture infrared wavelengths for monitoring vegetation. The use of infrared imagery is very common in satellite remote sensing, while most current near sensing studies are limited to visible wavelengths only. The use of infrared-visible observations is theoretically superior over the use of just visible observation due to the strong contrast between infrared and visible reflection of vegetation, the high correlation of the three visible bands and the possibilities to use spectral indices like the Normalized Difference Vegetation Index.This paper presents two experiments: the first study compares infrared modified and true color cameras to detect seasonal development of understory plants species in a forest; the second is aimed at evaluation of spectrometer and camera data collected during a laboratory plant stress experiment. The main goal of the experiments is to evaluate the utility of infrared modified cameras for the monitoring of plant health and phenology.Results show that infrared converted cameras perform less than standard color cameras in a monitoring setting. Comparison of the infrared camera response to spectrometer data points at limits in dynamic range, and poor band separation as the main weaknesses of converted consumer cameras. Our results support the use of standard color cameras as simple and affordable tools for the monitoring of plant stress and phenology.

Edge-preserving interpolation of depth data exploiting color information

The extraction of depth information associated to dynamic scenes is an intriguing topic, because of its perspective role in many applications, including free viewpoint and 3D video systems. Time-of-flight (ToF) range cameras allow for the acquisition of depth maps at video rate, but they are characterized by a limited resolution, specially if compared with standard color cameras. This paper presents a super-resolution method for depth maps that exploits the side information from a standard color camera: the proposed method uses a segmented version of the high-resolution color image acquired by the color camera in order to identify the main objects in the scene and a novel surface prediction scheme in order to interpolate the depth samples provided by the ToF camera. Effective solutions are provided for critical issues such as the joint calibration between the two devices and the unreliability of the acquired data. Experimental results on both synthetic and real-world scenes have shown how the proposed method allows to obtain a more accurate interpolation with respect to standard interpolation approaches and state-of-the-art joint depth and color interpolation schemes.

Application of day and night digital photographs for estimating maize biophysical characteristics

In this study, an inexpensive camera-observation system called the Crop Phenology Recording System (CPRS), which consists of a standard digital color camera (RGB cam) and a modified near-infrared (NIR) digital camera (NIR cam), was applied to estimate green leaf area index (LAI), total LAI, green leaf biomass and total dry biomass of stalks and leaves of maize. The CPRS was installed for the 2009 growing season over a rainfed maize field at the University of Nebraska-Lincoln Agricultural Research and Development Center near Mead, NE, USA. The vegetation indices called Visible Atmospherically Resistant Index (VARI) and two green–red–blue (2g–r–b) were calculated from day-time RGB images taken by the standard commercially-available camera. The other vegetation index called Night-time Relative Brightness Index in NIR (NRBINIR) was calculated from night-time flash NIR images taken by the modified digital camera on which a NIR band-pass filter was attached. Sampling inspections were conducted to measure bio-physical parameters of maize in the same experimental field. The vegetation indices were compared with the biophysical parameters for a whole growing season. The VARI was found to accurately estimate green LAI (R2 = 0.99) and green leaf biomass (R2 = 0.98), as well as track seasonal changes in maize green vegetation fraction. The 2g–r–b was able to accurately estimate total LAI (R2 = 0.97). The NRBINIR showed the highest accuracy in estimation of the total dry biomass weight of the stalks and leaves (R2 = 0.99). The results show that the camera-observation system has potential for the remote assessment of maize biophysical parameters at low cost.

6D scan registration using depth-interpolated local image features

This paper describes a novel registration approach that is based on a combination of visual and 3D range information. To identify correspondences, local visual features obtained from images of a standard color camera are compared and the depth of matching features (and their position covariance) is determined from the range measurements of a 3D laser scanner. The matched depth-interpolated image features allow one to apply registration with known correspondences. We compare several ICP variants in this paper and suggest an extension that considers the spatial distance between matching features to eliminate false correspondences. Experimental results are presented in both outdoor and indoor environments. In addition to pair-wise registration, we also propose a global registration method that registers all scan poses simultaneously.

Qualitative assessment of 3D flows using a flow visualisation technique

A three-dimensional flow visualisation technique has been developed and used to measure the three mean velocity components, the associated turbulence levels and the kinetic energy of turbulence. The experiment involved an inclined 45° circular jet in a rectangular enclosure. The technique makes use of three light sheets of different colour and a standard CCD colour camera interfaced to a video recorder and a PC. The analysis algorithm relies on determining the length and colour of the resulting tri-colour streaklines to provide rapid, field-wide, qualitative and quantitative velocity information which is not subject to ambiguity stemming from varying particle concentration fields. Characteristic results show good agreement with 3D-PIV data. The purpose of the experiment was the calibration of the technique rather than the study of the flow, hence the results reported should only be considered as an indicative for the study of the jet flow.

Dual-LED Imaging System for Secure and Robust Fingerprint Detection

Counterfeiting is an important issue in unattended fingerprint identification. To help prevent this, we have proposed to look at the finger color changes during an input action. When a finger is pressed against an optical fingerprint sensor, it is gradually deformed and the color of the fingerprint images changes. This is due to the blood movements inside the finger. If the extent of this color change exceeds a certain threshold, the finger is judged to be real. We start by describing the spectral changes in light scattered by the pressed finger. As the pressure applied to the finger increases, the relative intensity of the red portion of the spectrum decreases. Second, we discuss a dual-LED imaging system based on scattered light detection. Here, two LEDs with peak emissions at 530nm and 630nm are mounted on the edge of a plastic plate serving as a light-guide. Light scattered by a finger on the light-guide is captured by a standard color camera. Using eight types of replicas and live fingers of 42 volunteers, we have acquired series of images during an input action. For each finger, three input actions have been recorded. Analysis of this database of 150 input trials shows that our imaging system can differentiate the live fingers and the replicas clearly. We have investigated signal extraction process for a reliable differentiation. The imaging system is adequately robust against temperature fluctuation in a finger and it can be used under illumination of 3050 lux.

Process control of laser conduction welding by thermal imaging measurement with a color camera

Conduction welding offers an alternative to keyhole welding. Compared with keyhole welding, it is an intrinsically stable process because vaporization phenomena are minimal. However, as with keyhole welding, an on-line process-monitoring system is advantageous for quality assurance to maintain the required penetration depth, which in conduction welding is more sensitive to changes in heat sinking. The maximum penetration is obtained when the surface temperature is just below the boiling point, and so we normally wish to maintain the temperature at this level. We describe a two-color optical system that we have developed for real-time temperature profile measurement of the conduction weld pool. The key feature of the system is the use of a complementary metal-oxide semiconductor standard color camera leading to a simplified low-cost optical setup. We present and discuss the real-time temperature measurement and control performance of the system when a defocused beam from a high power Nd:YAG laser is used on 5 mm thick stainless steel workpieces.

Determination of mixing time by colourimetric diagnosis?Application to a new mixing system

A method for determining mixing times in a transparent stirred vessel for highly viscous fluids has been developed. The method involves a chemical decolourization with two acid-base indicators and a colourimetric diagnosis of digital images, deduced from a video captured during the mixing process. The advantages of the method developed are:(1) there is no disturbance of the fluid flow, and (2) measuring and determination of the degree of homogeneity is relatively effortless and cheap since it requires only a standard colour camera and commercial software for digital image analysis. This paper describes protocols for carrying this method out. In addition, examples of experimental results obtained for a standard impeller (of the helical ribbon agitator type) and a new piece of mixing equipment are given and discussed.