Professional Documents
Culture Documents
AI Projectlist 2021
AI Projectlist 2021
(Machine Learning, Artificial Intelligence, Data Science, Image Processing, Video Processing, Networking)
3. Early detection Deep Learning Type Convolution Neural Network Architecture for Multiclass
Classification of Alzheimer’s Disease
Alzheimer’s disease (AD) is one of the common medical issues that the world is facing today. This disease has
a high prevalence of memory loss and cognitive decline primarily in the elderly. At present, there is no
specific treatment for this disease, but it is thought that identification of it at an early stage can help to
manage it in a better way. Several studies used machine learning (ML) approaches for AD diagnosis and
classification. In this study, we considered the Open Access Series of Imaging Studies-3 (OASIS-3) dataset
with 2,168 Magnetic Resonance Imaging (MRI) images of patients with very mild to different stages of
cognitive decline. We applied deep learning-based convolution neural networks (CNN) which are well-known
approaches for diagnosis-based studies. The model training was done by 70% of images and applied 10-fold
cross-validation to validate the model. The developed architecture model has successfully classified the
different stages of dementia images and achieved 83.3% accuracy which is higher than other traditional
classification techniques like support vectors and logistic regression.
5. Fire and Gun Violence based Anomaly Detection System Using Deep Neural Networks
Real-time object detection to improve surveillance methods is one of the sought-after applications of
Convolutional Neural Networks (CNNs). This research work has approached the detection of fire and
handguns in areas monitored by cameras. Home fires, industrial explosions, and wildfires are a huge problem
that cause adverse effects on the environment. Gun violence and mass shootings are also on the rise in certain
parts of the world. Such incidents are time sensitive and can cause a huge loss to life and property. Hence, the
proposed work has built a deep learning model based on the YOLOv3 algorithm that processes a video frame-
by-frame to detect such anomalies in real-time and generate an alert for the concerned authorities. The final
model has a validation loss of 0.2864, with a detection rate of 45 frames per second and has been
benchmarked on datasets like IMFDB, UGR, and FireNet with accuracies of 89.3%, 82.6% and 86.5%
respectively. Experimental result satisfies the goal of the proposed model and also shows a fast detection rate
that can be deployed indoor as well as outdoors.
7. Human Activity Recognition Based on Graman Angular Field and Deep Convolutional Neural
Network
With the development of the Internet of things (IoT) and wearable devices, the sensor-based human activity
recognition (HAR) has attracted more and more attentions from researchers due to its outstanding
characteristics of convenience and privacy. Meanwhile, deep learning algorithms can extract high-
dimensional features automatically, which makes it possible to achieve the end-to-end learning. Especially
the convolutional neural network (CNN) has been widely used in the field of computer vision, while the
influence of environmental background, camera shielding, and other factors are the biggest challenges to it.
However, the sensor-based HAR can circumvent these problems well. Two improved HAR methods based on
deep CNN are proposed in this paper. Firstly, through the multi-dilated kernel residual (Mdk-Res) module, a
new improved deep CNN network Mdk-ResNet is proposed, which extracts the features among sampling
points with different intervals. Furthermore, the Fusion-Mdk-ResNet is adopted to process and fuse data
collected by different sensors automatically. The comparative experiments are conducted on three public
activity datasets, which are WISDM, UCI HAR and OPPORTUNITY. The optimal results are obtained by using
the indexes such as accuracy, precision, recall and F-measure, which verifies the effectiveness of the proposed
methods.
fake. It enhance the security of image frameworks, by adding assessment in a fast, user-friendly, and non-
intrusive manner, through the use of image quality assessment.
9. Detection of skin cancer using deep learning and image processing technique
Disease comparable to dermatologists and could empower lifesaving and quick judgments, even external the
clinic by means of establishment of applications on cell phones. As far as anyone is concerned, at present,
there is no audit of the ebb and flow work in this examination region. This investigation presents the main
orderly audit of the cutting edge research on characterizing skin sores with CNNs. We limit our audit to skin
injury classifiers. Specifically, strategies that apply a CNN just for division or for the order of dermoscopic
designs are not considered here. Moreover, this investigation talks about why the equivalence of the
introduced methodology is exceptionally troublesome and which difficulties should be tended to later on. We
looked through Google Scholar, PubMed, Medline, Science Direct, and Web of Science information bases for
orderly surveys and unique examination articles distributed in English. Just papers that announced adequate
logical procedures are remembered for this survey. We discovered 13 papers that grouped skin sores
utilizing CNNs. On a basic level, characterization strategies can be separated by three standards. Approaches
that utilization a CNN previously prepared through another enormous dataset and afterwards streamline its
boundaries to the grouping of skin sores are the most widely recognized ones utilized and they show the best
exhibition with the presently accessible restricted datasets. CNN's show is superior as cutting edge skin sore
classifiers. Shockingly, it is hard to think about various arrangement strategies since certain methodologies
utilize nonpublic datasets for preparing as well as testing, consequently making reproducibility troublesome.
Future distributions should utilize openly accessible benchmarks and completely reveal techniques utilized
for preparing to permit equivalence.
11. A Deep Neural Framework for Continuous Sign Language Recognition by Iterative Training
In this paper author describing new deep learning architecture to recognize Sign Language and this technique
is the combination of deep learning CNN (convolution neural networks), RNN (Recurrent Neural Network
Model called BiLSTM (Bidirectional Neural Network) to improve sign language recognition. Existing
technique were dependent on HMM (Hidden Markov Model) and this model was not accurate and to
overcome from this issue author has introduce CNN+BILSTM model where CNN layers can be used for
features extraction and to train deep learning model and then BILSTM can be used as the sequence learning
model to recognize possible next sequence of sign language. This model train itself iteratively with new
sequences predicted or recognize by BILSTM model.
12. Automatic Motorcyclist Helmet Rule Violation Detection using Tensorflow & Keras in OpenCV.
Motorcycle accidents have been hastily growing throughout the years in several countries because road
safety is often neglected by riders worldwide leading to accidents and deaths. To address this issue, most
countries have laws which mandate the use of helmets for two-wheeler riders so, it is very important for
motorcyclists to understand the risks of riding without a helmet. Riders who do not wear helmets are at
greatest risk of suffering a traumatic brain injury; if they met with an accident without protection, the head is
susceptible to a harrowing impact in an accident. In India, there is a rule that mandate helmet only for riders
but not even for passengers. Anyone may suffer from accident or head injuries whom are using motorcycle
without helmet. It should be mandatory for everyone to wear helmet; even for children. So, to mandate this
we have developed a system which is based on Tensorflow & Keras in the field of Computer Vision. System is
able to detect whether motorcyclists wear helmet or not even at real time. If anyone of them is present with
no helmet then system will precisely observed the situation and declare the rule violations. The system can be
implemented in malls, offices, marts, school and college that only allows people to enter the premises only
after detecting helmet with automated barrier. It will definitely affect the use of helmet that will save humans
life at all.
14. Automatic Detection of Diabetic Retinopathy: A Review on Datasets, Methods and Evaluation
Metrics
Diabetic retinopathy (DR) is a fast-spreading disease across the globe, which is caused by diabetes. The DR
may lead the diabetic patients to complete vision loss. In this scenario, early identification of DR is more
essential to recover the eyesight and provide help for timely treatment. The detection of DR can be manually
performed by ophthalmologists and can also be done by an automated system. In the manual system, analysis
and explanation of retinal fundus images need ophthalmologists, which is a timeconsuming and very
expensive task, but in the automated system, arti_cial intelligence is used to perform an imperative role in the
area of ophthalmology and speci_cally in the early detection of diabetic retinopathy over the traditional
detection approaches. Recently, numerous advanced studies related to the identi_cation of DR have been
reported. This paper presents a detailed review of the detection of DR with three major aspects; retinal
datasets, DR detection methods, and performance evaluation metrics. Furthermore, this study also covers the
author's observations and provides future directions in the _eld of diabetic retinopathy to overcome the
research challenges for the research community.
15. Plant Leaf Diseases Detection and Classification Using Image Processing and Deep Learning
Techniques.
Agricultural products are the primary need for every country. If plants are infected by diseases, this impacts
the country's agricultural production and its economic resources. This paper presents a system that is used to
classify and detect plant leaf diseases using deep learning techniques. The used images were obtained from
(Plant Village dataset) website. In our work, we have taken specific types of plants; include tomatoes, pepper,
and potatoes, as they are the most common types of plants in the world and in Iraq in particular. This Data Set
contains 20636 images of plants and their diseases. In our proposed system, we used the convolutional
neural network (CNN), through which plant leaf diseases are classified, 15 classes were classified, including
12 classes for diseases of different plants that were detected, such as bacteria, fungi, etc., and 3 classes for
healthy leaves. As a result, we obtained excellent accuracy in training and testing, we have got an accuracy of
(98.29%) for training, and (98.029%) for testing for all data set that were used.
18. Efficient Masked Face Recognition Method during the COVID-19 Pandemic.
The COVID-19 is an unparalleled crisis leading to huge number of casualties and security problems. In order
to reduce the spread of coronavirus, people often wear masks to protect themselves. This makes the face
recognition a very dicult task since certain parts of the face are hidden. A primary focus of researchers during
the ongoing coronavirus pandemic is to come up with suggestions to handle this problem through rapid and
ecient solutions. In this paper, we propose a reliable method based on discard masked region and deep
learning based features in order to address the problem of masked face recognition process. The rst step is to
discard the masked face region. Next, we apply a pre-trained deep Convolutional neural networks (CNN) to
extract the best features from the obtained regions (mostly eyes and forehead regions). Finally, the Bag-of-
features paradigm is applied on the feature maps of the last convolutional layer in order to quantize them and
to get a slight representation comparing to the fully connected layer of classical CNN. Finally, MLP is applied
for the classication process. Experimental results on Real-World-Masked-Face-Dataset show high recognition
performance
19. Age and Gender Prediction using Deep Convolutional Neural Networks
Age and gender identification have become a major part of the network, security and care. It has a common
use in age specific content access for children. Social media uses it in delivering layered ads and marketing to
extend it’s a reach. Face recognition has developed to a great extent that we have to map it further in getting
more useful results having different approaches. In this paper, we propose deep CNN to improve age and
gender predication from significant results can be obtained and a significant improvement can be seen in
various tasks such as face recognition. A simple convolutional network architecture is proposed to make a
noticeable improvement in this field using existing methods. Using deep CNN, model is trained to an extent
that accuracy of Age and Gender become 79% using HAAR Feature-based Cascade Classifiers is an
effectivemethod proposed by Paul Viola and Michael Jones. It is a machine learning based approach where a
cascade function is trained from a lot of positive and negative images. It is then used to detect objects in other
images.
20. Computer Vision for Attendance and Emotion Analysis in School Settings.
This paper presents facial detection and emotion analysis software developed by and for secondary students
and teachers. The goal is to provide a tool that reduces the time teachers spend taking attendance while also
collecting data that improves teaching practices. Disturbing current trends regarding school shootings
motivated the inclusion of emotion recognition so that teachers are able to better monitor students’
emotional states over time. This will be accomplished by providing teachers with early warning notifications
when a student significantly deviates in a negative way from their characteristic emotional profile. This
project was designed to save teachers time, help teachers better address student mental health needs, and
motivate students and teachers to learn more computer science, computer vision, and machine learning as
they use and modify the code in their own classrooms. Important takeaways from initial test results are that
increasing training images increases the accuracy of the recognition software, and the farther away a face is
from the camera, the higher the chances are that the face will be incorrectly recognized.
chest radiographs were initially screened for quality control by removing all low quality or unreadable scans.
The diagnoses for the images were then graded by two expert physicians before being cleared for training the
AI system. In order to account for any grading errors, the evaluation set was also checked by a third expert.