Open Source Computer Vision Library Research Articles

OPEN SOURCE AND AFFORDABLE TERRESTRIAL LASER SCANNER

Abstract. This paper explores the development of an affordable terrestrial laser scanner (TLS) system based on low-cost components and open source software. This paper enumerates components implemented and integrated in both hardware and software aspects. Primarily, the proposed system depends on Mid-40 LiDAR sensor and utilizes Point Cloud Library (PCL), Open Source Computer Vision Library (OpenCV) and Robotic Operating System (ROS) for system operation and data processing. Furthermore, this paper discusses sensor calibration and fusion methods available and implemented in the processing pipeline. The goal of this work is to produce an open source package of software system and hardware designs of a robust version of the TLS system that can be shared and easily reproduced by individuals and communities for their applications.

Hill Climb Game Play with Webcam Using OpenCV

Abstract: Computer vision is a process by which we can understand how the images and videos are stored and manipulated, also it helps in the process of retrieving data from either images or videos. Computer Vision is part of Artificial Intelligence. Computer-Vision plays a major role in Autonomous cars, Object detections, robotics, object tracking, etc. OpenCV (Open Source Computer Vision Library) is an open source computer vision and machine learning software library. OpenCV was built to provide a common infrastructure for computer vision applications and to accelerate the use of machine perception in the commercial products. It comes with a highly improved deep learning (dnn ) module. This module now supports a number of deep learning frameworks, including Caffe, TensorFlow, and Torch/PyTorch. This does allow us to take our models trained using dedicated deep learning libraries/tools and then efficiently use them directly inside our OpenCV scripts. MediaPipe is a framework mainly used for building audio, video, or any time series data. With the help of the MediaPipe framework, we can build very impressive pipelines for different media processing functions like Multi-hand Tracking, Face Detection, Object Detection and Tracking, etc.

OpenCV Using on a Single Board Computer for Incorrect Facemask-Wearing Detection and Capturing

OpenCV (Open Source Computer Vision Library) is a software library intended for real-time dynamic image processing, created by Intel. In this study, the library will be used to detect the face, nose and mouth. Furthermore, the system is equipped with the knowledge that if the mouth and nose or one of them is detected, then the face has not used the mask correctly and the system records the face. The system is supported by an image capture device in the form of a camcorder, a processing device in the form of a single board computer, such as a Raspberry Pi and a display device in the form of a monitor. And the result is that the system is able to make a decision whether the face is wearing a mask correctly or not. By means of labeling around the face in the form of red angular lines, if not properly use the mask. The success rate is 88.4% using detector parameters, namely: scale factor = 1.1 for all face, nose and mouth object libraries.

Primary Contacts Identification for COVID-19 Carriers from Surveillance Videos

COVID-19 (Coronavirus disease of 2019) is caused by SARS-CoV2 (Severe Acute Respiratory Syndrome Coronavirus 2) and it was first diagnosed in December 2019 in China. As of 25th Aug 2021, there are 165 million confirmed COVID-19 positive cases and 4.4 million deaths globally. As of today, though there are approved COVID-19 vaccine candidates only 4 billion doses have been administered. Until 100% of the population is safe, no one is safe. Even though these vaccines can provide protection against getting seriously ill and dying from the disease, it does not provide 100% protection from getting infected and passing it on to others. The more the virus spreads; it has more opportunity to mutate. So, it is mandatory to follow all precautions like maintaining social distance, wearing mask, washing hands frequently irrespective of whether a person is vaccinated or not. To prevent spread of the virus, contact tracing based on social distance also becomes equally important. The work proposes a solution that can help with contact tracing/identification, knowing the infected persons recent travel history (even within the city) for few days before being assessed positive. While the person would be able to give the known contacts with whom he/she has interacted with, he/she will not be aware of who all were in proximity if he/she had been in public places. The proposed solution is to get the CCTV (Closed-Circuit Television) video clips from those public places for the specific date and time and identify the people who were in proximity—i.e., not followed the safe distance to the infected person. The approach uses YOLO V3 (You Only Look Once) which uses darknet framework for people detection. Once the infected person is located from the video frames, the distance from that person to the other people in the frame is found, to check if there is a violation of social distance guideline. If there is, then the people violating the distance are extracted and identified using Facial detection and recognition algorithms. Two different solutions for Face detection and Recognition are implemented and results compared—Dlib based models and OpenCV (Open Source Computer Vision Library) based models. The solutions were studied for two different CCTV footages and the results for Dlib based models were better than OpenCV based models for the studied videos.

Face mask detection with temperature check

The coronavirus (COVID-19) pandemic is causing a worldwide health catastrophe, so according to the World Health Organization (WHO), wearing masks in public is an effective safety method. The COVID-19 pandemic has forced governments around the world to impose quarantines to prevent transmission of the virus. According to reports, wearing masks in public does reduce the threat of transmission of the virus. An efficient and cost-effective way to use Artificial Intelligence (AI) to create a secure environment in a manufacturing environment. A hybrid model for using a deep and classic face mask detection device will be proposed. The face mask detection dataset includes the mask, and without mask photos, it uses the Open-Source Computer Vision Library (OpenCV) to detect faces in real-time from the stay circulation through the webcam. It uses the dataset to build a computer vision COVID-19 face mask detector using Python, OpenCV, TensorFlow, and Keras. Using computer vision and deep learning, the goal is to understand whether a character in a picture or video stream is wearing a mask or not using computer vision and deep learning.

Real time Drowsy Driver Detection using Polynomial Kernel based Support Vector Machine

Abstract: Casualty increases from road accidents day by day. There are so many reasons that accident causes and mostly due to human errors. Driver drowsiness is one of them. A small drowsiness may turn it into a big accident that resulted heavy casualties. If any of the system automatically detects the driver’s drowsiness and alert at real time may secure many lives. Drowsiness can be recognized by different situations such as by opening full mouth, by closing both the eyes and a combination of both. This may advised not to drive at drowsy state. There are various techniques through which drowsiness can be detected at real time but accuracy matters. OpenCV is a highly utilized open source computer vision library through which facial features can be recognized effectively. Polynomial kernel based support vector machine (SVM) is an advanced classification technique through which drowsiness can be classified from face. SVM is advanced machine learning approach through which linear and non-linear data can be classified with higher level of accuracy. System pertained 96.17 % of accuracy. Polynomial kernel is useful for non-linear data separation. Here system classifies the expressional features of face and result accordingly for drowsiness detection. Keywords: Support Vector Machine (SVM), OpenCV, Machine Learning, Non-Linear SVM Model, Drowsiness Detection, Face Detection, Computer Vision.

Emotion based video player

One's work can be done efficiently only if their mood is good. Emotions is the index of mood. Here model capture one's image as an input, predict their mood and play a video of opposite genre as an output, in order to change their mood, which is the main goal of this project. Hence taking them through an emotional roller coaster. The solution makes use of CNN (convolutional neutral networks) for detecting one's mood. It uses OpenCV (open-source computer vision library) in-order to get user's image using their respective web camera. It is done by importing modules like web-browser and requests in-order to get access to YouTube to play videos accordingly. The average accuracy rate of the system has increased to 98.53 percent. Eight primary emotion classes have been effectively classified by the method. As a result, the proposed strategy has been shown to be effective in recognizing emotions.

Research on Face Recognition Based on Improved Convolutional Neural Network Using Raspberry Pi

In this paper, a face recognition system has been designed using Raspberry Pi based on convolution neural network technology. Convolution neural network that using OpenCV (open source computer vision library) to realize real-time face recognition and feature extraction. The difficult and important problem in deep learning network is that it could be over fitting very easily. In order to alleviate the over fitting phenomenon and improve the accuracy of face recognition, the dropout technology of convolutional neural network had been employed to improve the recognition accuracy and ensure the robustness of face recognition. The results of experiments show that our methods has the advantages in the computer time and recognition rate for the low construction cost, simple structure and strong portability. The accuracy of face detection for static and dynamic images using Raspberry Pi is above 85% in our testament.

ГЕОДЕЗІЯ, КАРТОГРАФІЯ І АЕРОФОТОЗНІМАННЯ

The aim of the work is to develop an automated measuring system in a mechanical gyrocompass with the help of specially developed hardware and software in order to facilitate the operation of the device and minimize observer errors. The developed complex provides automation only for the time method, as for the method of the turning point it is necessary to constantly contact the motion screw of the total station. The project is based on an integrated system, the hardware part of which contains a single-board computer, camera, and lens. The main software is a developed motion recognition algorithm with the help of image processing. This algorithm was created using the Python programming language and the open-source computer vision library OpenCV. With the help of the hardware, a video image of the gyroscope's reference scale is obtained, and with the help of the software, the moving light indicator and its position relative to the scale are identified in this image. The result of the study is a functioning automatic measurement system, which determines the value of the azimuth of the direction with the same accuracy as manual measurements. The system is controlled remotely via a computer and wi-fi network. To test the system, a series of automatic and manual measurements were performed simultaneously at the same point for the same direction. Based on the results obtained, it can be stated that the accuracy of the system is within the limits specified by the manufacturer of the device for manual measurements. The application of computer vision technology, namely the tracking of a moving object in the image for gyroscopic measurements can give a significant impetus to the development of automation systems for a wide range of measuring instruments, which in turn can improve the accuracy of measurement results. The developed system can be used together with the Gyromax AK-2M gyrocompass of GeoMessTechnik for carrying out automated measurements, training of new operators. With the help of the developed model, it is possible to avoid gross errors of the observer, to facilitate the measurement process which will not demand the constant presence of the operator near the device. In some dangerous conditions, this is a significant advantage.

Facial Recognition in Public Areas

The security of information nowadays is very significant and difficult, so there are a number of ways to improve security. Especially in public areas like airports, railway stations, Universities, ATMs, etc. and security cameras are presently common in these areas. So, in this paper, we are presenting how Facial recognition can be used in public areas like airports, toll gates, offices, etc. We are comparing or matching a face of a person who we want to detect, with the video which is recorded through CCTV. There are certain algorithms to detect faces from video like through HAAR cascades, eigenface, fisher face, etc. open-source computer vision library is used for facial recognition.

Automatic Social Distance Monitoring system using Deep Learning Algorithms

In recent months, the worldwide outbreak of Corona Virus Disease (COVID-19) has created a panic situation for all human beings irrespective of the geographic location and economic status of the countries. The communal spread of the deadly virus can be prevented by maintaining the social distance between people and the wearing of face masks. This paper provides a prominent solution for monitoring using the Open-Source Computer Vision Library (Open-CV) and Deep Learning Algorithm to course people in public areas and to avoid crowding. This proposed project can be implemented using images taken from the Closed Circuit Television (CCTV). The distance between the people is detected with help of an object detection algorithm. The Euclidean distance is calculated and compared with the standard distance given. The persons with enough distance will be indicated in the green-colored bounding box and others with a red-colored bounding box. The total number of violations is also shown at the output feed. This project provides a solution to eradicate the spread of the deadly virus to an extent with minimum human effort and reduced risk of infection to the local athourities.

Computer vision reimages rigor of APPT pain site and surface area measurement

The Adolescent Pediatric Pain Tool (APPT) contains a body outline diagram (BOD) to facilitate self-report of pain sites and surface area. Inter-rater reliability for manually coded BOD drawings is only fair to moderate. Irregular drawings are prone to manual measurement error. The clinical utility, trending, and interpretability of BOD drawings are significant limitations of APPT. A rigorous method to code BODs and extract data for further analysis is needed. Aim: To digitize BOD drawings and using computer vision improve accuracy of BOD measures to facilitate further analysis. Open source computer vision library, OpenCV, was used to digitize BODs from adolescents with scoliosis (n=42) or pectus excavatum (n=24) obtained preoperatively, daily postoperatively during hospitalization, and at postoperative visits. Pain sites and areas were drawn by adolescents on paper BODs, then scanned and cleaned through grayscaling, median blurring, and thresholding techniques to reduce noise. Scanned images were aligned with reference diagram using homography matrix. Differences between images and reference were extracted. Identified body parts and surface areas were labeled for further analysis. The model was able to extract BOD drawings and identify pain sites that corresponded to APPT scoring template. Correct classification rate for each body location for each patient was 97.2%. Overall sensitivity was 95.5% and specificity 97.3%. This computer vision method of coding BODs is accessible, convenient, and provides better sensitivity and specificity for pain site and surface area than manual coding. Extracted data can be further analyzed for trends and modeled for pain trajectory prediction. While the use of digital applications to prospectively collect BOD data may overcome current APPT limitations, loss of historic data, expense, data access, and electronic medical record (EMR) integration present new limitations. Paper BODs are convenient, inexpensive, easily scanned into EMRs, and now easily analyzed.

A Tree Ring Measurement Method Based on Error Correction in Digital Image of Stem Analysis Disk

Stem analysis is an essential aspect in forestry investigation and forest management, as it is a primary method to study the growth law of trees. Stem analysis requires measuring the width and number of tree rings to ensure the accurate measurement, expand applicable tree species, and reduce operation cost. This study explores the use of Open Source Computer Vision Library (Open CV) to measure the ring radius of analytic wood disk digital images, and establish a regression equation of ring radius based on image geometric distortion correction. Here, a digital camera was used to photograph the stem disks’ tree rings to obtain digital images. The images were preprocessed with Open CV to measure the disk’s annual ring radius. The error correction model based on the least-square polynomial fitting method was established for digital image geometric distortion correction. Finally, a regression equation for tree ring radius based on the error correction model was established. Through the above steps, click the intersection point between the radius line and each ring to get the pixel distance from the ring to the pith, then the size of ring radius can be calculated by the regression equation of ring radius. The study’s method was used to measure the digital image of the Chinese fir stem disk and compare it with the actual value. The results showed that the maximum error of this method was 0.15 cm, the average error was 0.04 cm, and the average detection accuracy reached 99.34%, which met the requirements for measuring the tree ring radius by stem disk analysis. This method is simple, accurate, and suitable for coniferous and broad-leaved species, which allows researchers to analyze tree ring radius measurement, and is of great significance for analyzing the tree growth process.

Geleneksel Kuru Kaymakların Bazı Kimyasal, Tekstürel, Mikrobiyolojik ve Duyusal Özellikleri

Kuru kaymak, geleneksel kaymağın tandırda ve/veya gölgede kurutulması ile üretilen bir süt ürünüdür. Bu çalışmada, Erzurum ve Nevşehir piyasalarından toplanan Kuru kaymak örneklerinde (10 adet) bazı fiziksel, kimyasal, tekstürel, mikrobiyolojik ve duyusal özelliklerin belirlenmesi amaçlanmıştır. Ayrıca örneklerin görüntüleri, OpenCV (Open Source Computer Vision Library) görüntü işleme kütüphanesi ve C++ dili kullanılarak analiz edilmiştir. Kuru kaymak örneklerinde laktik asit (%) değerinin %0,46-1,21 aralığında ve pH değerinin 6,19-6,83 aralığında değiştiği saptanmıştır. Örneklerin ortalama kurumadde, yağ ve kül içerikleri sırasıyla %71,44, %42,25 ve %2,70 olarak tespit edilmiştir. Yapılan renk ölçümleri ile ortalama L* değerinin 77,42, a* değerinin 2,13 ve b* değerinin 17,44 olduğu belirlenmiştir. Kuru kaymak örneklerinin ortalama sertlik değerinin 1761,37 N, esneklik değerinin 4,30, iç yapışkanlık değerinin 0,64, sakızımsılık değerinin 689,41 N, çiğnenebilirlik değerinin 1212,00 N ve elastikiyet değerinin 0,60 olduğu tespit edilmiştir. Görüntü işleme analizi sonucuna göre örneklerin gözeneklilik durumu %12,27-29,84 aralığında değişmektedir. Mikrobiyolojik değerlendirme sonucunda örneklerde ortalama 3,94 log kob/g maya-küf ve 3,97 log kob/g koliform grubu bakterilerin bulunduğu belirlenmiştir. Ayrıca 4 örnekte Escherichia coli tespit edilmiştir. Görünüş, yapı ve koku parametreleri örneklerde sırasıyla ortalama 5,77, 5,75 ve 5,49 olarak tespit edilmiştir. Bu çalışmada, birbirinden farklı özelliklerde Kuru kaymak örneklerinin olduğu anlaşılmaktadır. Unutulmak üzere olan bu geleneksel süt ürününün özelliklerinin optimize edilmesi, standart ve hijyenik koşullarda üretilmesinin gerekli olduğu düşünülmektedir.

Facial Emotion Recognition In Real-time Using Deep Convolution Networks

The objective of this study is to present a detection method associated to automatic live facial expression identification by analysing the frontal-face image and predicting the most accurate expression out of the seven major countenances. Since our behaviour is strongly correlated to our emotions, the facial expressions and body gestures may act as a noteworthy source of non-verbal communication that may tell about the state of an individual. The interest in emotional computing through facial expressions has increased as it has wide application in industries, market and medical field. The doctors may be helped through facial expression recognition, may be by online machine monitoring system or sometimes when any patient is not able to communicate verbally. This study presents a model for facial recognition that permits disturbances to apprehend information from their surroundings in real-time, thus improving the classification process. The work has been implemented using Python IDLE (3.7) and Open Source Computer Vision Library (Open-CV2). The study proves to provide accurate and precise results.

STML (Sketch to Markup Language)

“Web development encompasses many stages, which involve designing the website done by making a wireframe first, after which the code for the frontend and backend is appropriately written over many iterations until the developers arrive at a fully functional and satisfying website for their need. We see how every website starts with the basic idea and structure of the website that can be described using a wireframe. It shows the basic elements and gives developers an idea about the structure of the website. It is given to the developers to create the boilerplate code that places all the elements into place accordingly. The task of converting the wireframe to HTML code is tedious and time-consuming. At present, the boilerplate code is mostly written manually. Currently, the users have to write the HTML code for structuring the elements on the webpage, which leads to a lot of redundant work and users investing their precious time into it. Generally, where the structure of web pages is the same, users would tend to maintain a copy of the pre-existing boilerplate code for reusability. In most cases, the boilerplate code differs for web pages despite the code for the HTML elements being the same. In such situations, the work becomes redundant. However, this process can be automated to ease the world of the web developer community. To make this process effortless, we propose a machine learning model. It will be trained to identify specific symbols and shapes as elements, recognize texts from the wireframe. The model will take in wireframe images as input through the web application. The idea is to process input, recognize each element in the wireframe using the Open Source Computer vision library (OpenCV). Identified elements will each have a corresponding code in the backend. Once the elements on the wireframe are identified, the corresponding code is written into an HTML file. The user gets the HTML file as an output. The creation of the boilerplate code for a website becomes much less time-consuming, which gives developers the freedom to test out many different designs and layouts before opting for the one that best suits their needs.”

Whole-Brain Image Analysis and Anatomical Atlas 3D Generation Using MagellanMapper.

MagellanMapper is a software suite designed for visual inspection and end-to-end automated processing of large-volume, 3D brain imaging datasets in a memory-efficient manner. The rapidly growing number of large-volume, high-resolution datasets necessitates visualization of raw data at both macro- and microscopic levels to assess the quality of data, as well as automated processing to quantify data in an unbiased manner for comparison across a large number of samples. To facilitate these analyses, MagellanMapper provides both a graphical user interface for manual inspection and a command-line interface for automated image processing. At the macroscopic level, the graphical interface allows researchers to view full volumetric images simultaneously in each dimension and to annotate anatomical label placements. At the microscopic level, researchers can inspect regions of interest at high resolution to build ground truth data of cellular locations such as nuclei positions. Using the command-line interface, researchers can automate cell detection across volumetric images, refine anatomical atlas labels to fit underlying histology, register these atlases to sample images, and perform statistical analyses by anatomical region. MagellanMapper leverages established open-source computer vision libraries and is itself open source and freely available for download and extension. © 2020 Wiley Periodicals LLC. Basic Protocol 1: MagellanMapper installation Alternate Protocol: Alternative methods for MagellanMapper installation Basic Protocol 2: Import image files into MagellanMapper Basic Protocol 3: Region of interest visualization and annotation Basic Protocol 4: Explore an atlas along all three dimensions and register to a sample brain Basic Protocol 5: Automated 3D anatomical atlas construction Basic Protocol 6: Whole-tissue cell detection and quantification by anatomical label Support Protocol: Import a tiled microscopy image in proprietary format into MagellanMapper.

Analyzing the Impact of Iridescence Strength on Animal Shape Recognition Using Computer Vision

Iridescence, commonly found in nature, is the quality of a material to change colors depending on its interaction with light. For example, certain species of beetles distract predators with their iridescent exoskeletons that flash different colors depending on the angle the predator looks at them. Iridescent flowers, however, attract pollinators through their noticeable changes in color. Thus, it seems that iridescence can serve two contrasting purposes: concealment and attraction. The purpose of this project is to test whether different strengths in iridescence can impact its recognizability to animals and insects. 
 Three different iridescent diffraction grating films were experimented with: 500 lines/mm, 532 lines/mm, and 1000 lines/mm. Matte paper was used as a control. Four circular epoxy resin disks (8 cm diameter) and four square epoxy resin disks (7 cm side length) were created. Each diffraction grating film type and the matte paper was glued onto one disk of each shape. Photos of each disk were then taken at angles of 30, 45, 60, and 90 degrees in a photo booth and processed through a computer vision algorithm, the Ramer-Douglas-Peucker algorithm, to perform contour approximation.
 The results gathered cannot conclude that the strengths of iridescence have an impact on shape recognition and detectability due to limitations and potential errors. Nonetheless, the study can suggest that the specific usage of the Ramer-Douglas-Peucker algorithm for edge and area detection through OpenCV (an open-source computer vision library) is inadequate in imitating animal and insect visual systems when analyzing iridescence.

Computer Vision-based Reader for analogue Energy/Water Meters in low-cost embedded System: a Case Study in an Office Building in Scotland

Implementation of cost-effective energy conservation measures (ECMs) is expected to generate up to 18% of carbon emissions reductions in office buildings. In order to determine adequate ECMs for a specific building, operational data is required. However, buildings generally lack operational data in the form of time series that can limit a breath of analysis required for determining adequate ECMs. Energy time-series data is commonly lacking in the UK due to uneven availability of smart meters (heat, gas, water), security restrictions in Energy Information Systems (EIS) and building management systems (BMS), restrictions and costs associated for automated reporting from utility companies, etc. This work presents a non-intrusive computer vision-based reader to generate energy readings at 10-minute resolution using a Raspberry-Pi, a traditional webcam and an LED light. OpenCV, an open source computer vision library, is used to detect and interpret numeric values from a heat meter, which are in turn uploaded to a cloud-based energy platform to create a complete operational data set enabling detailed analytics, fault detection and diagnostics (FDD) and model calibration. A case study of an office building in Scotland is presented. The building has a heat meter with no remote access capabilities. The accuracy of the method, i.e. the ability of the script to accurately derive the rate of change between readings, resulted on a 92% percent during a test done for 100 samples. Recommendations for accuracy improvements are included in the conclusions.

Precipitating Cloud Analysis during Extreme Rain Events over the Chao Phraya River Basin Based on the Developed Radar Mosaic Products

During rainy season, Tropical Cyclones (TCs) regularly provide heavy rain over Indochina. In addition, the monsoon trough also produces rainfall to support agricultural activities across the area. However, the extreme rainfall events usually occur during the seasonal monsoon march. Understanding the precipitating cloud characteristics produced by the extreme rainfall events could alleviate the destruction of natural resources over Indochina. The variation in precipitating clouds can be analyzed using fine spatio-temporal resolution data, such as ground-based radar data from multiple radar stations. Since the Chao Phraya river basin is the most important basin in Thailand, this study is based on radar images collected from the five installed radar stations of the Thai Meteorological Department over this basin. The analysis of the extreme rain events during rainy season in 2018 was carried out using the developed mosaic method. Python and Open Source Computer Vision Library (OpenCV) were used in the development of mosaic products to focus on characteristics of precipitating clouds. Beam-blockage fraction was analyzed for each radar station and surrounding terrain using an open source library for processing weather radar data (wradlib) (Heistermann et al., 2013) with our developed beam-blockage analysis program. The severity of blocked beam is found over the northern mountain region of study area. The precipitating clouds were detected and their parameters extracted using the ellipse fitting technique. The analysis of these clouds revealed that, among other cases, Sontihn TC was a severely destructive event based on the storm parameters of both stratiform and convective clouds. The number and length of the convective storms during Sontihn TC was the most severe when compared to other events. It has been found that the diurnal cycle of the convective cloud shows their patterns by following solar heating surface over the study area for all of the selected events.