Real-Time Raindrop Detection Based on Deep Learning Algorithm

Layer-wise 3D surface morphology information is critical for the quality monitoring and control of additive manufacturing (AM) processes. However, most of the existing 3D scan technologies are either contact or time consuming, which are not capable of obtaining the 3D surface morphology data in a real-time manner during the process. Therefore, the objective of this study is to achieve real-time 3D surface data acquisition in AM, which is achieved by a supervised deep learning-based image analysis approach. The key idea of this proposed method is to capture the correlation between 2D image and 3D point cloud, and then quantify this relationship by using a deep learning algorithm, namely, convolutional neural network (CNN). To validate the effectiveness and efficiency of the proposed method, both simulation and real-world case studies were performed. The results demonstrate that this method has strong potential to be applied for real-time surface morphology measurement in AM, as well as other advanced manufacturing processes.

Download Full-text

Real Time LDR Data Prediction using IoT and Deep Learning Algorithm

10.46532/978-81-950008-1-4_033 ◽

2020 ◽

pp. 158-161

Author(s):

Chandraprabha S ◽

Pradeepkumar G ◽

Dineshkumar Ponnusamy ◽

Saranya M D ◽

Satheesh Kumar S ◽

...

Keyword(s):

Deep Learning ◽

Light Intensity ◽

Real Time ◽

Short Term Memory ◽

Learning Algorithm ◽

Time Interval ◽

Deep Learning Algorithm ◽

Data Prediction ◽

Solar Plant

This paper outfits artificial intelligence based real time LDR data which is implemented in various applications like indoor lightning, and places where enormous amount of heat is produced, agriculture to increase the crop yield, Solar plant for solar irradiance Tracking. For forecasting the LDR information. The system uses a sensor that can measure the light intensity by means of LDR. The data acquired from sensors are posted in an Adafruit cloud for every two seconds time interval using Node MCU ESP8266 module. The data is also presented on adafruit dashboard for observing sensor variables. A Long short-term memory is used for setting up the deep learning. LSTM module uses the recorded historical data from adafruit cloud which is paired with Node MCU in order to obtain the real-time long-term time series sensor variables that is measured in terms of light intensity. Data is extracted from the cloud for processing the data analytics later the deep learning model is implemented in order to predict future light intensity values.

Download Full-text

Indoor and Outdoor Fire Localization Using YOLO Algorithm

Journal of Physics Conference Series ◽

10.1088/1742-6596/2114/1/012067 ◽

2021 ◽

Vol 2114 (1) ◽

pp. 012067

Author(s):

Ruba R. Nori ◽

Rabah N. Farhan ◽

Safaa Hussein Abed

Keyword(s):

Deep Learning ◽

Real Time ◽

Learning Algorithm ◽

Binary Classification ◽

Learning Algorithms ◽

Fire Detection ◽

Time Data ◽

Deep Learning Algorithm ◽

Positive Elements ◽

Image Dataset

Abstract Novel algorithm for fire detection has been introduced. CNN based System localization of fire for real time applications was proposed. Deep learning algorithms shows excellent results in a way that it accuracy reaches very high accuracy for fire image dataset. Yolo is a superior deep learning algorithm that is capable of detect and localize fires in real time. The luck of image dataset force us to limit the system in binary classification test. Proposed model was tested on dataset gathered from the internet. In this article, we built an automated alert system integrating multiple sensors and state-of-the art deep learning algorithms, which have a limited number of false positive elements and which provide our prototype robot with reasonable accuracy in real-time data and as little as possible to track and record fire events as soon as possible.

Download Full-text

Real-Time Driver Fatigue Detection from ECG Using Deep Learning Algorithm

Design Science and Innovation - Ergonomics for Improved Productivity ◽

10.1007/978-981-15-9054-2_71 ◽

2021 ◽

pp. 615-621

Author(s):

V. Anjaly Cherian ◽

Rahul Bhardwaj ◽

Venkatesh Balasubramanian

Keyword(s):

Deep Learning ◽

Real Time ◽

Learning Algorithm ◽

Driver Fatigue ◽

Deep Learning Algorithm ◽

Fatigue Detection

Download Full-text

Real-Time Road Monitoring Using Deep Learning Algorithm Deployed on IoT Devices

10.1201/9781003140351-15 ◽

2021 ◽

pp. 137-147

Author(s):

Nilay Nishant ◽

Ashish Maharjan ◽

Dibyajyoti Chutia ◽

P. L. N. Raju ◽

Ashis Pradhan

Keyword(s):

Deep Learning ◽

Real Time ◽

Learning Algorithm ◽

Deep Learning Algorithm ◽

Iot Devices

Download Full-text

Digital Twin For Offshore Pipeline Corrosion Monitoring: A Deep Learning Approach

10.4043/31296-ms ◽

2021 ◽

Author(s):

Subrata Bhowmik

Keyword(s):

Deep Learning ◽

Real Time ◽

Processing Time ◽

Learning Algorithm ◽

Corrosion Monitoring ◽

Digital Twin ◽

Pipeline Inspection ◽

Deep Learning Algorithm ◽

Inspection Data ◽

Pipeline Corrosion

Abstract Pipeline corrosion is a major identified threat in the offshore oil and gas industry. In this paper, a novel computer vision-based digital twin concept for real-time corrosion inspection is proposed. The Convolution Neural Network (CNN) algorithm is used for the automated corrosion identification and classification from the ROV images and In-Line Inspection data. Predictive and prescriptive maintenance strategies are recommended based on the corrosion assessment through the digital twin. A Deep-learning Image processing model is developed based on the pipeline inspection images and In-Line Inspection images from some previous inspection data sets. During the corrosion monitoring through pipeline inspection, the digital twin system would be able to gather data and, at the same time, process and analyze the collected data. The analyzed data can be used to classify the corrosion type and determine the actions to be taken (develop predictive and prescriptive maintenance strategy). Convolution Neural Network, a well known Deep Learning algorithm, is used in the Tensorflow framework with Keras in the backend is used in the digital twin for corrosion inspection. CNN algorithm will first detect the corrosion and then the type of corrosion based on image classification. The deep-learning network training is done using 4000 images taken from the inspection video frames from a subsea pipeline inspection using ROV. The performances of both the methods are compared based on result accuracy as well as processing time. Deep Learning algorithm, CNN has approximately 81% accuracy for correctly identifying the corrosion and classify them based on severity through image classification. The processing time for the deep-learning method is significantly faster, and the digital twin generates the predictive or prescriptive strategy based on the inspection result in real-time. Deep-learning based digital twin for Corrosion inspection significantly improve current corrosion identification and reduce the overall time for offshore inspection. The inspection data loss due to the communication interference during real-time assessment can be eliminated using the digital twin. The image data can recover the required features based on other features available through the previous inspection. Furthermore, the system can adapt to the unrefined environment, making the proposed system robust and useful for other detection applications. The digital twin makes a recommended decision based on an expert system database during the real-time inspection. The complete corrosion monitoring process is performed in real-time on a cloud-based digital twin. The proposed pipeline corrosion inspection digital twin based on the CNN method will significantly reduce the overall maintenance cost and improve the efficiency of the corrosion monitoring system.

Download Full-text

Vision-Based Deep Learning Algorithm for Detecting Potholes

10.20944/preprints202102.0569.v1 ◽

2021 ◽

Author(s):

Kanushka Gajjar ◽

Theo van Niekerk ◽

Thomas Wilm ◽

Paolo Mercorelli

Keyword(s):

Deep Learning ◽

Real Time ◽

Autonomous Vehicles ◽

Learning Algorithm ◽

Detection System ◽

Detection Methods ◽

Deep Learning Algorithm ◽

Learning Techniques ◽

Pothole Detection ◽

Adequate Data

Potholes on roads pose a major threat to motorists and autonomous vehicles. Driving over a pothole has the potential to cause serious damage to a vehicle, which in turn may result in fatal accidents. Currently, many pothole detection methods exist. However, these methods do not utilize deep learning techniques to detect a pothole in real-time, determine the location thereof and display its location on a map. The success of determining an effective pothole detection method, which includes the aforementioned deep learning techniques, is dependent on acquiring a large amount of data, including images of potholes. Once adequate data had been gathered, the images were processed and annotated. The next step was to determine which deep learning algorithms could be utilized. Three different models, including Faster R-CNN, SSD and YOLOv3 were trained on the custom dataset containing images of potholes to determine which network produces the best results for real-time detection. It was revealed that YOLOv3 produced the most accurate results and performed the best in real-time, with an average detection time of only 0.836s per image. The final results revealed that a real-time pothole detection system, integrated with a cloud and maps service, can be created to allow drivers to avoid potholes.

Download Full-text

Real-Time Detection and Location of Potted Flowers Based on a ZED Camera and a YOLO V4-Tiny Deep Learning Algorithm

Horticulturae ◽

10.3390/horticulturae8010021 ◽

2021 ◽

Vol 8 (1) ◽

pp. 21

Author(s):

Jizhang Wang ◽

Zhiheng Gao ◽

Yun Zhang ◽

Jing Zhou ◽

Jianzhi Wu ◽

...

Keyword(s):

Deep Learning ◽

Real Time ◽

Point Cloud ◽

Learning Algorithm ◽

Absolute Error ◽

Frame Rate ◽

3D Point Cloud ◽

The Real ◽

Deep Learning Algorithm ◽

The Mean

In order to realize the real-time and accurate detection of potted flowers on benches, in this paper we propose a method based on the ZED 2 stereo camera and the YOLO V4-Tiny deep learning algorithm for potted flower detection and location. First, an automatic detection model of flowers was established based on the YOLO V4-Tiny convolutional neural network (CNN) model, and the center points on the pixel plane of the flowers were obtained according to the prediction box. Then, the real-time 3D point cloud information obtained by the ZED 2 camera was used to calculate the actual position of the flowers. The test results showed that the mean average precision (MAP) and recall rate of the training model was 89.72% and 80%, respectively, and the real-time average detection frame rate of the model deployed under Jetson TX2 was 16 FPS. The results of the occlusion experiment showed that when the canopy overlap ratio between the two flowers is more than 10%, the recognition accuracy will be affected. The mean absolute error of the flower center location based on 3D point cloud information of the ZED 2 camera was 18.1 mm, and the maximum locating error of the flower center was 25.8 mm under different light radiation conditions. The method in this paper establishes the relationship between the detection target of flowers and the actual spatial location, which has reference significance for the machinery and automatic management of potted flowers on benches.

Download Full-text

Real Time Crime Detection Using Deep Learning Algorithm

10.1109/icscan53069.2021.9526393 ◽

2021 ◽

Author(s):

P. Sivakumar ◽

Jayabalaguru. V ◽

Ramsugumar. R ◽

Kalaisriram. S

Keyword(s):

Deep Learning ◽

Real Time ◽

Learning Algorithm ◽

Deep Learning Algorithm ◽

Crime Detection

Download Full-text