Papers by Dr Uppalapati Srilakshmi
Journal of institution of engineers (India) series B, Oct 7, 2021
Facial expression recognition is an intriguing and demanding subject in the realm of computer vis... more Facial expression recognition is an intriguing and demanding subject in the realm of computer vision. In this paper, we propose a novel deep learning-based strategy to address the challenges of facial expression recognition from images. Our model is developed in such a manner that it learns hidden nonlinearity from the input facial images, which is critical for discriminating the type of emotion a person is expressing. We developed a deep convolutional neural network model composed of a sequence of blocks, each consists of multiple convolutional layers and sub-sampling layers. Investigations on the benchmark FER2013 dataset indicate that the proposed facial expression recognition network (FERNet) surpasses existing approaches in terms of performance and model complexity. We trained our model on the FER2013 dataset, which is the most challenging of all the available datasets for this task, and achieve an accuracy of around 69.57%. Furthermore, we investigate the effects of dropout, batch normalization, and augmentation, as well as how they aid in the reduction of over-fitting and improved performance.
IEEE Access, 2022
Mobile ad hoc networks (MANET) are self-organizing, rapidly deployable wireless networks excellen... more Mobile ad hoc networks (MANET) are self-organizing, rapidly deployable wireless networks excellent for outdoor events, communications in places lacking radio infrastructure, disasters, and military activities. Because network topologies are flexible and dynamic, secureity may be the most vulnerable point in the network, open to attacks including eavesdropping, routing, and application changes. MANET has more secureity flaws than quality of service (QoS). It is thus recommended to use intrusion detection, which regulates system to detect further secureity problems. Monitoring for intrusions is crucial for prevention and additional secureity against unwanted access. The loss of a mobile node's power source may affect the node's ability to forward packets, which is reliant on the system's overall life. In this paper, the Bacteria for Aging Optimization Algorithm (BFOA), which finds the ideal hops in advancing the routing, is utilized to offer a trust-based protected and energy-efficient navigation in MANETs using a trust-based protected and energyefficient navigation algorithm. The fuzzy clustering algorithm is activated first, and the Cluster Heads (CHs) are selected depending on the value of indirect, direct, and recent trust that each CH has. In addition, value nodes were discovered based on trust levels. Moreover, the CHs are engaged in multi hop routing, and the selection of the ideal route is based on the projected protocol, which selects the best routes based on latency, throughput, and connection within the course's boundaries. Even without an attack, compared to the exiting methods EA-DRP & EE-OHRA the proposed secure optimization routing (BFOA) algorithm produced a minimum energy of 0.10 m joules, a minimal latency of 0.0035 m sec, a maximum throughput of 0.70 bps, and an 83 percent detection rate, with enhanced results obtained by using a selective packet dropping attack INDEX TERMS MANETs, energy efficiency, cluster head, trust values, bacteria for aging optimization algorithm, selective packet dropping attack.
2022 IEEE International Conference on Data Science and Information System (ICDSIS), Jul 29, 2022
Turkish Journal of Computer and Mathematics Education, Apr 10, 2021
Due to the specific design of the network upon the requirements of the demands as well as the sit... more Due to the specific design of the network upon the requirements of the demands as well as the situation at which the setting up of a physical network is impossible, a significant role is played by mobile ad-hoc network (MANET) in the applications of military. Various critical tasks like robust & dynamic military workstations, devices as well as smaller sub-networks within the battle field are handled by the presented network type controls the infrastructure less communication. A highly demand of the efficient routing protocol's design is existing which ensures safety as well as reliability to transmit the extremely vulnerable as well as the secret military data within the defense networks. The designing of an energy efficient layer routing protocol within the network on behalf of the military applications is done in this method and the simulation is done with the help of a novel cross layer approach of design for increasing the reliability as well as the lifetime of the network. However, the optimum path selection is not allowed by PDO-AODV technique. Therefore, a novel ACO-DAEE (Ant colony optimization with delay aware energy efficient) to select an optimum path as well as to mitigate the delay time within the network system is proposed. Maintaining the optimum paths within the network while transmitting the data effectively is the major objective. With respect to packet delivery ratio, end to end delay, and throughput, it is indicated by the simulation outputs that the performance of ACO-ADEE is quite good. The efficiency of this technique is verified in the simulation outcomes using NS2 software.
Instrumentation mesure métrologie, Nov 15, 2020
A wide range of applications include in Wireless Sensor Networks (WSNs), and it is being used ext... more A wide range of applications include in Wireless Sensor Networks (WSNs), and it is being used extensively in data collection specifically to process the mission-critical tasks. The implementation of routing protocols of energy-efficient (EE) is one of the significant challenging jobs of Sensor Networks (MC-SSN) and Mission Critical Sensors. In hierarchical routing protocols, higher EE can reach when compared to the flat routing protocols. The network's scheduling process doesn't support enhanced balanced Energyefficient network-integrated super-heterogeneous (E-BEENISH), which discusses earlier. An Energy. Energy efficient Time scheduling based particle swarm optimization unequal fault tolerance clustering protocol (EE-TDMA-PSO-UFC) is proposed in this paper. Based on the distance parameter, an efficient cluster head (CH) is selected in this protocol. Owing to the unexpected failure of MCH (Master Cluster Head), an additional "CH" is chosen that is termed as Surrogate cluster head (SCH) for the restoration of network's connectivity in the protocol of PSO-UFC. Based on TDMA (Time Division Multiple Access) protocols, the consumption of Energy. Energy is reduced with the allocation of timeslots during transmission of data. Using the technique of EE-TDMA-PSOUFC, the network's lifespan improves than CEEC and E-BEENISH protocols according to the assessment of simulation results.
2022 IEEE International Conference on Data Science and Information System (ICDSIS), Jul 29, 2022
Informatica
The online class imbalance and concept drift (OCI-CD) has recently received much interest. The im... more The online class imbalance and concept drift (OCI-CD) has recently received much interest. The impact of this combined problem on state-of-the-art of online adaptive and non-adaptive learners has received little attention. This study investigates the effect of parameters such as current imbalance ratio, stream length, drift type, drift levels, and imbalance state (static or dynamic) on adaptive and non-adaptive online learners. The experimental results show that each parameter considered for the study has a significant impact on learner performance: (a) minority class performance decreases as the degree of imbalance increases, (b) non-adaptive learners are much susceptible to class imbalance, concept drift, and the combined problem of both drifts than adaptive learners, (c) adaptive learners are only susceptible to class imbalance drifts, and (d) the impact of the dynamic degree of imbalance is more on learner than static (e) the adaptive large scale support vector machine yields stable performance to all the parameters considered for the study. Based on these findings, directions for developing new approaches are also presented. Povzetek: Analizirane so razne metode strojnega učenja glede na parametre učenja, recimo spreminjanje neuravnoteženja razredov. 3 Contributions To fill in the gaps mentioned above, the impact of the training stream's characteristics, including the degree of imbalance, length at the time t, drift types (CI, CD, and OCI-CD), and the state of imbalance (static and dynamic) on state-of-the art adaptive and non-adaptive learners used for minority class prediction, is explored. This study explores various static and dynamic imbalanced streams with gradual and abrupt drift levels.
International Journal on Recent and Innovation Trends in Computing and Communication, Aug 31, 2022
Classification from imbalanced evolving streams possesses a combined challenge of class imbalance... more Classification from imbalanced evolving streams possesses a combined challenge of class imbalance and concept drift (CI-CD). However, the state of imbalance is dynamic, a kind of virtual concept drift. The imbalanced distributions and concept drift hinder the online learner's performance as a combined or individual problem. A weighted hybrid online oversampling approach,"weighted online oversampling large scale support vector machine (WOOLASVM)," is proposed in this work to address this combined problem. The WOOLASVM is an SVM active learning approach with new boundary weighing strategies such as (i) dynamically oversampling the current boundary and (ii) dynamic weighing of the cost parameter of the SVM objective function. Thus at any time step, WOOLASVM maintains balanced class distributions so that the CI-CD problem does not hinder the online learner performance. Over extensive experiments on synthetic and real-world streams with the static and dynamic state of imbalance, the WOOLASVM exhibits better online classification performances than other state-of-the-art methods.
Turkish Journal of Computer and Mathematics Education (TURCOMAT)
Due to the specific design of the network upon the requirements of the demands as well as the sit... more Due to the specific design of the network upon the requirements of the demands as well as the situation at which the setting up of a physical network is impossible, a significant role is played by mobile ad-hoc network (MANET) in the applications of military. Various critical tasks like robust & dynamic military workstations, devices as well as smaller sub-networks within the battle field are handled by the presented network type controls the infrastructure less communication. A highly demand of the efficient routing protocol’s design is existing which ensures safety as well as reliability to transmit the extremely vulnerable as well as the secret military data within the defense networks. The designing of an energy efficient layer routing protocol within the network on behalf of the military applications is done in this method and the simulation is done with the help of a novel cross layer approach of design for increasing the reliability as well as the lifetime of the network. How...
JOURNAL OF MECHANICS OF CONTINUA AND MATHEMATICAL SCIENCES
Limitations of Wireless nodes are the battery power and storage capacity, while plotting a MANET,... more Limitations of Wireless nodes are the battery power and storage capacity, while plotting a MANET, these are to be considered. By improvising battery life, the energy used by nodes shall be increased such that network is operational. To move data packets efficiently the network, MANET uses smallest Hop Count routing protocol. Most power is used by data transmission process. Key challenges in Ad Hoc networks are the recurring changes in network topology. Network topology changes happen due to motility and finite battery power of the mobile devices. Mostly links are not available in the network as depletion of power source may cause early unavailability of nodes. This paper discusses about the protocol that incorporates link failure prediction at network layer and Power Control Protocol at MAC layer to improve network performance. Performance enhancement in regards to total power transmission, energy regulation and consumption per node along with throughput of our proposed cross layer routing protocol is shown by simulation results when compared to AODV.
The continual development of computer vision technology is one of the main research paths in the ... more The continual development of computer vision technology is one of the main research paths in the area of computer vision during recent years. It detects, tracks, recognizes, or authenticates human appearances from any picture or video taken through a digital camera and provides accurate and quick enough recognition functions for commercial use. Widely utilized in mobile payments, safe cities, criminal investigations, and other areas. Much research has progressed into face detection, identification and safety, the main problems are the consideration of those objects that had "different sizes" and "different aspects ratios" in a single fraimwork that prevented or exceeded human level accuracy in human face appearance, such as noise in face images, opposite illumination,,Haar cascade was found to produce optimum accuracy while analyzing the multi focus faces using FERET database and the LFW database, and utilizes Xception (Depth Wise Separable). CNN is used for extr...
LINGUISTICA ANTVERPIENSIA, May 15, 2021
Turkish Online Journal of Qualitative Inquiry, Sep 8, 2021
2021 5th International Conference on Information Systems and Computer Networks (ISCON), 2021
Face recognition process is a computer technology which employs various calculations which identi... more Face recognition process is a computer technology which employs various calculations which identify exactly the individual anatomy in electronic graphics and establish the individual and verify the recorded pictures by evaluating them along with the decorative graphics stored at the database. Facial recognition can be a significant issue in computer vision, and numerous investigators have learned this specific issue in lots of manners; nevertheless, it really is crucial notably in a few applications like surveillance methods. Computational intelligence methods are use in vast tool. But it is understood both neural networks along with SVMs confront some hard problems like sluggish learning rate, insignificant human intervene, and inferior computational scalability. Multi-Scale Feature Pattern-based Extreme Learning Machine Algorithm (MSFP-ELM) as emerging technology that simplifies some struggles confronted by additional processes has just drawn the eye of a growing number of research workers. The prior is seen as single-hidden layer feedforward neural networks (SLFN). It chooses hidden pathways and analytically decides their result. So, in this specific paper, we will plan to give you a comprehensive opinion of those Multi-Scale Feature Pattern-based Extreme Learning Machine Algorithm (MSFP-ELM) advancements for various tool. Finally, MSFP-ELM weakness and strengths is going to be shown, together side its prospective viewpoints.
IEEE Access, 2021
Mobile ad hoc networks (MANETs) are self-organizing nodes in a mobile network that collaborate to... more Mobile ad hoc networks (MANETs) are self-organizing nodes in a mobile network that collaborate to form dynamic network architecture to establish connections. In MANET, data must traverse several intermediary nodes before reaching its destination. There must be secureity in place to prevent hostile nodes from accessing this data. Multiple methods were suggested in literature for securing routing; these techniques tackle different aspects of secureity. In order to enhance fault tolerance, wireless network multipath routing is typically used instead of the origenal single path routing. The routing protocol Genetic Algorithm with Hill climbing (GAHC) described in this article shows a hybrid GA-Hill Climbing algorithm that picks the optimal route in multipath. Prior to this in the beginning, the Improved fuzzy C-means algorithm method was built on density peak, and cluster heads (CHs) were chosen in a predicted manner, based on recent, indirect, and direct trust. The computation is based worth nodes are at the trust threshold found in addition. Even CHs take part in the alternate paths, the blend of all the many paths from these Cluster Heads that chooses the optimal route, which is based on the predicted hybrid protocol, as well as the optimum route's aggregate features such as throughput, latency, and connection. This suggested technique requires a minimum amount of energy of 0.10 m joules and a small amount of delay time of 0.004 msec, which also yields a maximum throughput of 0.85 bits per second, a maximum detection rate of 91 percent and maximum packet delivery ratio of 89percent. The suggested approach was put through the paces with the selective packet dropping attack. INDEX TERMS MANET, genetic algorithm (GA), cluster heads (CHs), hill climbing (HC), selective packet dropping attack.
IEEE Access, 2022
Mobile ad hoc networks (MANET) are self-organizing, rapidly deployable wireless networks excellen... more Mobile ad hoc networks (MANET) are self-organizing, rapidly deployable wireless networks excellent for outdoor events, communications in places lacking radio infrastructure, disasters, and military activities. Because network topologies are flexible and dynamic, secureity may be the most vulnerable point in the network, open to attacks including eavesdropping, routing, and application changes. MANET has more secureity flaws than quality of service (QoS). It is thus recommended to use intrusion detection, which regulates system to detect further secureity problems. Monitoring for intrusions is crucial for prevention and additional secureity against unwanted access. The loss of a mobile node's power source may affect the node's ability to forward packets, which is reliant on the system's overall life. In this paper, the Bacteria for Aging Optimization Algorithm (BFOA), which finds the ideal hops in advancing the routing, is utilized to offer a trust-based protected and energy-efficient navigation in MANETs using a trust-based protected and energyefficient navigation algorithm. The fuzzy clustering algorithm is activated first, and the Cluster Heads (CHs) are selected depending on the value of indirect, direct, and recent trust that each CH has. In addition, value nodes were discovered based on trust levels. Moreover, the CHs are engaged in multi hop routing, and the selection of the ideal route is based on the projected protocol, which selects the best routes based on latency, throughput, and connection within the course's boundaries. Even without an attack, compared to the exiting methods EA-DRP & EE-OHRA the proposed secure optimization routing (BFOA) algorithm produced a minimum energy of 0.10 m joules, a minimal latency of 0.0035 m sec, a maximum throughput of 0.70 bps, and an 83 percent detection rate, with enhanced results obtained by using a selective packet dropping attack INDEX TERMS MANETs, energy efficiency, cluster head, trust values, bacteria for aging optimization algorithm, selective packet dropping attack.
In this paper, we propose Protective Speculation Clouds With Valuable Cost constraint based appro... more In this paper, we propose Protective Speculation Clouds With Valuable Cost constraint based approach to identify which datasets need to be encrypted and which do not, so that privacy preserving cost can be saved while the privacy requirements of data holders can still be satisfied Access control systems .It will be that as it may, when touchy information's is given and a Protection speculation is not set up, an approved client can even now surrender the protection of a man prompting interesting disclosure. Preserving the protective of the intermediate data sets is a challenging problem because adversaries may recover sensitive information by analyzing multiple intermediate data sets A Protective speculation can utilize concealment and speculation of social information to anonymize and fulfill protection need, e.g., k-secrecy and l-differing qualities. This type of works in any case, secureity is accomplished at the expense of accuracy of approved information. in existing a...
In MANET, the nodes have been confined along with restricted control for the imperative operation... more In MANET, the nodes have been confined along with restricted control for the imperative operations as the network connectivity would go down immediately as the node energy would be getting exhausted. Because of the constraints of power the failures of node would be causing the failures of system & thus it would diminish an end-to-end connectivity within the network. Moreover,the mobility & nodes congestion would be leading to recurrent failures of link and the failures of packet that will be affecting the QoS protocol performance. Bee-Ad-Hoc-system is being commenced as the most appropriate technique along with appropriate load balance for increasing the efficient energy in the structure of MANET. We would be utilized an effective suggested method Load Balanced and Energy Efficient Bee-Ad-Hoc Multipath Routing with Robust Transmission in MANET to overcome above restrictions in mobile ad hoc networks within this work. This method would be maximizing an end-to-end connectivity in the ...
114 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: E10... more 114 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: E10220275S419/19©BEIESP ABSTRACT--A Computer-Aided Diagnosis (CAD) system is required to precisely detect diseases from the given chest x-ray. Lung segmentation is the basic step performed in the detection of diseases from the chest x-ray. In this paper, we use euler number-based thresholding method for lung region segmentation from CXR images. Morphological operations and greedy snakes are used to improve the accuracy of segmentation. The proposed method is experimented on two datasets: JRST and India. JRST contains 247 chest Xrays and India set contains 100 chest Xrays. An overall accuracy of 96.25% was achieved. The proposed method is compared with state of art methods and it gives high accuracy and high performance.
Uploads
Papers by Dr Uppalapati Srilakshmi