Artificial Intelligence in IoT and Cyborgization

The term “cyborg” refers to an organism that has recovered function or, more specifically, enhanced abilities as a result of the integration of artificial components or technology that depends on feedback, such as prostheses, artificial organs, implants, or, in some cases, wearable technology. Cyborgs are not the same as bionics, bio-robotics, or androids. Companies that produce cybernetic technology take part in a range of initiatives to encourage increased contact between people and computers. Collective intelligence may be supported or enabled by cyborg technologies. Cyborgs are used in various applications such as eye restoration, bodily networks, retinal implantation, vocal cords, artificial pancreas in medical field, military, sports, arts etc. The various case cyborg attempts, technical pros and cons are discussed briefly in this chapter.

The integration of IoT and AI has a potential to generate more efficient solutions and experiences. By combining the incoming data from IoT devices you can get more value from your network and improve your business. Artificial Intelligence is a critical part of IoT that makes intelligent network management and operations successful. From combination of these two modern technologies, smart gadgets will be created allowing businesses to make strategic decisions with least percentage of error. The applications of IoT are very wide such as virtual reality, mixed reality, augmented reality, use of it in healthcare, agriculture, disaster management, waste management and much more. The integration of AI with IoT has made possible many smart products commonly available in the market such as smart locks, Green IQ smart garden hubs, SmartMat and smart controllers. We are already on our way to a cyborg civilization; a cyborg is an entity where technology and biology are integrally attached. This could upgrade the human limits and abilities of a human and by this one could become more immune to injuries and have a much higher intelligence as well.

The Internet of Things (IoT) would contain a severe, well organized, and economical and communication effect in our everyday life. Links in IoT channels usually controlled by resources, where cyber-attacks are more likely. Extensive works have proposed to access security and secret issues on IoT channels to address these problems. However, the new characteristics of IoT links are not sufficient to link the top security concerns of IoT systems to present descriptions. Machine Learning (ML) and Deep Learning (DL) methods could give more knowledge of IoT devices that could help overcome different previous security issues. In this chapter, we properly debated security specifications and present security solutions for IoT systems. Then, we provide in-depth of the present ML and DL methods related to additional safety in IoT systems.

Huge changes as happened to the contemporary world by IOT means Internet of things based technology after it’s discover in the field of computer and internet. In this project bus-bar can be protected from over current condition. In today’s world, the. technological trend of implementing Smart technologies, fostered by emergence of Cloud computing and Internet of things, led to a transfiguration of ordinary devices also tend to transcend and become smart, and consequently offer improved fault-detection and protection, remote monitoring and event notifications. As a result, we may combine the two technologies to make the present power system more efficient and well-organized. IoT and smart grid combine to form a superb mix of two skills that will improve India's current power structure. In adding to that there will be many benefits of using this expertise. Many existing problems that are present in the conventional power grid structure can be solved. The motive of this paper is to improve the sharing out common situation.

Reconfigurable Holographic Surface (RHS) is one of the meta material radiation elements which are integrated with transceivers to generate electromagnetic waves, empowering an ultrathin edifice. RHS exploits the meta material radiation elements to hypothesis a holographic strategy based on the holographic interference principle. Each component has electrical control over the radiation amplitude of the occurrence electromagnetic surfs to produce anticipated guiding beams. A digital twin is a representation of a physical object made from sensor data in the digital realm. A digital twin can combine intangible sensor data with physical object data, such as the shape or position of the real device, to create a final dynamic digital twin. Digital twin includes both stationary and active information. In this chapter, we present a novel digital-twin framework for RHS-assisted wireless networks which is called as Environment-Twin (Env-Twin). The objective of the Env-Twin framework is to empower mechanization of optimal control at various coarseness. Deep learning techniques such as Convolution Neural Network (CNN) and long short-term memory architecture (LSTM) are used to build our model and studied its performance and sturdiness. In this chapter, we also inspect the nascent for a digital twin deep learning model is used to find the reflection co-efficient of reconfigurable holographic surface for the receiver locations without the need for channel estimation and beamforming algorithms.

Fishing is one of the main wellsprings of food and pay for practically all beach front terrains regardless of its geographical area in the earth. Since it plays a significant part to play in the Economy, adjoining nations having similar seas every now and again participate in questions with respect to responsibility for region. This has brought about issues for the anglers’ local area dwelling in the beach front districts of these nations. The angler’s intersection the boundaries and distinguishing proof of areas in the ocean is turning into a troublesome errand with existing gear give to the angler’s subsequently they cross the lines. In our everyday life, we catch wind of numerous Fishermen from tamilnadu being gotten and situate in Sri Lankan Navy guardianship. The ocean line among the nations isn’t effectively recognizable, which is the fundamental justification behind this offense. Besides, in instances of impending cataclysmic events, disappointment or postpone in telling concerned work force to empty outcomes in death toll for a huge scope. In this section, proposed a strategy that safeguards the anglers by logging their entrances and exits in the harbor utilizing an implanted framework, telling the country’s ocean line to them by utilizing the GPS and IOT. When they knowingly tries to cross the border the boat will automatically control the direction and come back to safe zone. Sharks are the fish, which will hit the fisher boats sometimes which puts the fishermen in a risky situation on the sea. Here shark detection system has developed with the help of image processing technology which will save the life of fishermen from shark attack.

To explore the application of Human–Computer Interaction (HCI) in industrial Digital Twins (DTs), the current application status of DTs in Intelligent Manufacturing (IM) and the HCI problem in human–computer assembly are explored; aiming at the Human Action Recognition (HAR) of machine perspective in human–computer assembly, it proposes the Human Pose Estimation (HPE) method based on improved HRNet and inroduces the attention mechanism to establish the SE_NewHRNet model for the optimization of HPE; in addition, it points out Adaptive Architecture of Deep Learning Based on Confrontation (ADLC), and performs case analysis for the performance verification of the model. The accuracy of SE_NewHRNet in \(\mathrm{A}{P}_{pose}\) indicator is 75.1%. Compared with other models, its network performance is improved to different extents, the number of parameters and calculation amount are lower, the Loss value decreases rapidly, and the decrease rate becomes slower after 40 iterations. In ADLC, the recognition accuracy of the branch model with three domain discrimination is the highest, reaching 86.50%; in most cases, ADLC performance is better in contrast to other models. Compared with the Wasserstein Generative Adversarial Networks (WGAN) model with the second comprehensive performance, the average accuracy of ADLC is 69.72, 9.24% higher than that of WGAN, and 27.67% higher than that of Source-only. Therefore, the proposed human recognition method performs better.

Deep Learning has been a revolutionary innovation in the field of medical imaging. The domains that once required hours of intense study for detection or classification of a disease has now exponentially reduced with the help of certain state of the art works of DL. In this work, we have proposed two types of classification procedures for the diseases Miosis and Mydriasis which unlike Anisocoria, extremely dilates or constricts both the pupils. This condition is popular among people with brain disease, traumatic brain injury and by medications like opioids. This is also common in the field of agriculture as one of the causes being direct eye contact with chemicals such as pesticides. The proposed approaches are based on Convolutional Neural Network and Hough transformation techniques for identifying the arbitrary shapes of iris and pupil. 

The subset of Artificial Intelligence is Deep Learning which is inspired from the arrangement and communication of neurons in the Brain. 2D-CNN models are used to propose a solution to the Separation Problem. Various previously trained Architecture in large databases such as VGG-16, V66-19, Inception V3, ResNet-50, DenseNet-201, etc. are available, which can be used, this technique being called as Transfer learning. The process of finding an object in a picture is called Object Detection. There may be more than one event of the same object or more than one type of object in the same image. Many of the initially proposed solutions to this problem depend entirely on the first proposed districts for procurement. Among the many proposed region proposal Algorithms, Complete Search, Selected Search, Slide Window and Edge Boxes are some of the algorithms most commonly used for object discovery. R-CNN, Fast R-CNN, Faster R-CNN, Mask R-CNN, YOLO, SSD, and Detectron can be said to be the most advanced and extensively used approaches, each with its own drawbacks and advantages of Object Detection. Splitting is the work of combining image pixels together based on a specific principle. The Mask R-CNN Algorithm also deals with the problem of fragmentation. This chapter is designed to elaborate on YOLO Object Detection Algorithms and SSD, a comprehensive application of Object Detection Algorithms in Clinical Image Analysis, and the use of YOLO Architecture to detect brain tumors from MRI Image results.

Nowadays, there are various expanding deadly illnesses that undermine both human wellbeing and life. Among the perilous infections, Diabetes is one of the most common illnesses in the world. A high blood glucose level, which has a major impact on human organs, is a guarantee of diabetes. There are currently 382 million diabetics worldwide, and by 2035, the International Diabetes Federation (IDA) projects that number to reach 592 million. Building expectation frames is essential when using Deep Learning algorithms. Deep learning, machine learning, and artificial intelligence are the burgeoning fields for creating computerized predictions and suggestions. In the current framework, diabetes was anticipated dependent on the places of the two peoples, breath examination, checking blood glucose level highlights on galvanic skin reaction utilizing different procedures, for example, a min–max work for standardization, Support Vector Machine (SVM), arbitrary backwoods, Artificial Neural Networks (ANN), and various AI calculations for prediction. The conventional classifiers, for example, SVM and Decision trees are likewise used to make a forecast model. In this proposed framework, we are thinking about numerous highlights, for example, Diabetic blood pressure (mm Hg), Skin thickness (mm), serum insulin (mu U/ml), Body Mass Index, Diabetes family capacity, and age (a long time). We utilize a choice emotionally supportive network for diabetes expectation dependent on a profound convolution neural organization. Pima Indian Diabetes Dataset is utilized to break down and perform experimentation and reproduction work. Deep Convolution Neural Networks (DCNN) is a reasonable model to separate more information highlights from the Diabetes Dataset that unite the organization. In this DNN model, Rectified Linear Activation Function (ReLu) is applied against the dataset for standardization. This actuation capacity can undoubtedly prepare the examples to accomplish high exactness that improves the framework execution. The exactness of the proposed technique estimated by utilizing accuracy, review, z-score, k-overlap cross-approval. This proposed DCNN strategy can address the test that isn’t overwhelmed by the current framework and improve the presentation of the expectation of diabetes model.

Deep learning is a combination of artificial intelligence and machine learning concept. It is greatly helpful to imitate human’s knowledge. Deep learning is a predictive model which analyses using statistics and data science. Targeted Treatment is mainly applied to cancer cells to target the affected cells in a precise manner. The treatment is given to particular type of cancer cells by specified drugs and other substance. Deep learning is used in targeted treatment to achieve perfect diagnosis particularly for assessing medical images. In medical field, especially in oncology field deep learning proves reliable and consistency diagnosis, accuracy and better efficiency.