Browse Items (16488 total)
Sort by:
-
Automatic smart attendance management system based on face recognition using deep learning /
Patent Number: 202141025155, Applicant: J Omana.
In this digital era, face recognition system plays a vital role in almost every sector. Face recognition is one of the mostly used biometrics. It can used for security, authentication, identification, and has got many more advantages. Furthermore, face recognition system can also be used for attendance marking in schools, colleges, offices, etc. The conventional method of calling name of each student is time consuming and there is always a chance of proxy attendance. -
Sexual harassment complaints across BSE100 companies: a longitudinal dataset (20132023)
Objective: This dataset offers empirical support for understanding the number of sexual harassment complaints across various industries and sectors, and for analysing patterns over a decade. The data was manually collected from the annual integrated report or business responsibility report of each company on the BSE100 list. This dataset can be utilised by the Indian Ministry of Women & Child Development and related stakeholders as a foundation for studying the impact of previous interventions and policy changes on complaint reporting rates. Additionally, it can also be used to analyse patterns and trends and give way to revisions to existing legal frameworks and improvements to grievance redressal mechanisms for sexual harassment complaints. Data description: Data includes descriptive statistics for all variables and compares complaints received between each year and the subsequent year, with a total of 10 year-to-year comparisons for the period: 20132024. The variables are: (1) List of BSE100 Companies, (2) Industry, (3) Sector, (4) Number of sexual harassment complaints received for the periods: 20132024, (5) Number of sexual harassment complaints pending during the periods: 20132024. The Author(s) 2025. -
IOT based plant disease detection using support vector machine algorithm /
Patent Number: 202021050705, Applicant: Dr. Chaitanya Singh.
Internet of Things integrated with image processing paves new ways for monitoring plant health and actuating response based on recent development of technologies. Early detection of plant diseases and its classification utilizes image processing along with analysis of environmental data which will help farmers to maximize their yield by producing healthy plants as plant diseases are eradicated at the initial stage. -
Machine learning based EEG signal processing for smart patient monitoring system /
Patent Number: 202141022214, Applicant: Dr. Ganesh Kumar R.
In the current pandemic situation, patients with critical diseases are lacking immediate care which would reduce the mortality rate. This invention focuses on continuous monitoring of patient™s EEG signals for occurrence of any seizures in brain signals. This system is designed using machine learning algorithm for resource optimization thereby implemented using VLSI technology. -
A system for regulated distributed deep learning in cloud and smart mobile devices /
Patent Number: 202121025112, Applicant: Supriya Prashant Diwan.
The system in the present invention keeps the private data locally in smartphones, shares trained parameters and builds a global consensus model. The feasibility and usability of the proposed system are evaluated by three experiments and related discussion. The experimental results show that the distributed deep learning system can reconstruct the behavior of centralized training. We also measure the cumulative network traffic in different scenarios and show that the partial parameter sharing strategy does not only preserve the performance of the trained model but also can reduce network traffic. -
Agriculture water purifying device /
Patent Number: 363685-001, Applicant: S Nithyananthi. -
Fingerprint based authentication model to protect the sim card of the mobile /
"Patent Number: 201941048500, Applicant: Dr. Gobi R.
Mobile becomes an essential part of our life, and it is also an inevitable medium for communication in the present era. It keeps us connected from mail access, travel booking, etc. The amount of personal information and sensitive information like username, password, and transaction details are routinely stored in the device. There is a possibility of information access when the thief steals the Mobile. It is a massive data loss for the mobile owner, especially private information. -
Web User Access Log Analytics Using Neural Learning, Regression and Logit Boost Clustering Techniques for Accurate User Behavioural Pattern Identification
Web Usage Mining (WUM), is the process of mining user behaviour patterns from huge log fles. Weblogs provide substantial input to learning the identity of an online user. Analysis of these patterns extracted from the weblog datasets is currently being explored by various researchers. Due to the recent advent of automation, mining patterns from weblogs are automated. These automated mining processes focus on browsing habits and usage patterns. To make this process of gathering better, there are many ways to look at how users act and put them into relevant groups.Identifying, detecting, and classifying features that demarcate specifc traits that are related is an important task. Conventional research is designed to discover web usage mining strategies through clustering and classifcation methods. However, there is a need to focus on and improve the accuracy of the prediction systems that classify acquired features to fgure out the patterns of web users. Deep learning methods are used to mine weblog data to improve accuracy and precision. To improve user behaviour pattern mining, a two-level clustering process is introduced as Ensemble Fuzzy K-Means with Logit Boost Clustering (EFK-LBC) technique to extract the weblog. In this technique, a preprocessing step is included to remove redundant data and choose reliable log fles. The Fuzzy-K means clustering technique is used to identify behavioural patterns exhibited by recurrent users. Finally, the Logit Boost Clustering method is introduced to the data,that help in generating a strong cluster. Clustering of web users frequent behavioural patterns using the Logit Boost ensemble technique helps the proposed EFK-LBC method to improve newlinethe accuracy up to 88% and reduce the clustering time by 20% compared with existing approaches. Though the proposed EFK-LBC technique performs better for user identifcation, the different initialization of clusters provides various fnal clustering results. -
Artificial Intelligence - Based Steganography Model for Social Media Data Set
Steganography, one of the data security mechanisms under our investigation, shields legitimate messages from hackers and spies by employing data hiding. Data protection is newlinecurrently the top priority due to the signifcant advancements in information technology due to high-security concerns. Traditional techniques for maintaining data confdentiality include steganography and cryptography; the distinction is that steganography does not naturally arouse suspicion, whereas cryptography does. Traditional linguistic steganographic methods suffer from limitations in automation, accuracy, and the volume of concealed text. The robustness and undetectability properties of these approaches also require improvement. Third-party vulnerability is often too high for conventional techniques to handle. Artifcial intelligence is increasingly replacing traditional model creation in steganography. Despite the fact that steganography ensures security, information sent over online social networks (OSN) is plainly not safe. Steganography along newlinewith encryption can make a difference with regard to privacy of information in transit. newlineThe research study aims to build algorithms or models and assess steganography s robustness, security, undetectability, and embedding ability. Two distinct types of data newlineconcealing employed for investigation: text and image. The results were encouraging newlinewhen we initially tested our Laplacian model using image steganography and compared newlinewith benchmark methods. The second experiment, which is based on AI, generates the cover text using secret information, examines the security and robustness of steganography. The study compared suggested text steganography model, 3-bit data concealing, with other existing techniques in order to ascertain the undetectability factor. The frst experiment used MATLAB tools, and the second used the markovify python module, RNN (Recurrent Neural networks), and the Huffman tree. Further format-based steganography methods utilized in the following experiment. -
Cr2C MXene quantum dots for selective detection of mercury ions
Mercury (Hg), a highly toxic environmental contaminant that presents significant ecological and health risks, even at low concentrations. There is an urgent need for precise and sensitive methods for detecting Hg2+ ions in the environment. In recent years, a new class of 2D layered materials, MXene, has gained enormous attention due to their unique properties, such as high surface area, oxidation resistance, thermal and chemical stability, electrical and thermal conductivity. This study presents the synthesis and characterization of Cr2C MXene quantum dots (MQDs) derived from Cr2CTx nanolayered MXene sheets via the probe-sonication method. The Cr2C-MQDs were characterized using XRD, FTIR, SEM-EDS mapping, and Zeta potential analysis. The vibrant green fluorescence material, Cr2C-MQDs, was investigated for Hg2+ detection, which exhibited high selectivity and stability with a limit of detection of 30.7 nM. The sensing mechanism is attributed to the strong affinity of Cr2C MXene quantum dots for Hg2+ ions. 2025 Elsevier B.V. -
Green is the new beautiful: exploring the role of sustainability in building brand equity in cosmetic brands
Purpose This study investigates how embracing environmental responsibility and eco-friendly innovation in cosmetic brands cultivates brand equity. It specifically explores the mediating roles of a green brand image and consumer trust in driving sustainable value creation through these initiatives. Design/methodology/approach A quantitative research design was employed using survey data from 326 online respondents, mainly young and educated females. A structured questionnaire with Likert-scale items adapted from validated studies measured environmental responsibility, eco-friendly product innovation, green brand image, consumer trust and buying engagement. Data analysis included reliability testing, factor analysis and structural equation modeling with bootstrapping to examine direct and mediating relationships. Findings The study results show that perceived environmental responsibility strongly predicts buying engagement and significantly improves green brand image and consumer trust. Eco-friendly product innovation positively influences consumer trust but does not directly strengthen green brand image, indicating that innovation alone is insufficient without clear communication. Green brand image and consumer trust act as key mediators influencing buying engagement. Originality/value This study presents an empirically validated framework explaining how perceived environmental responsibility and eco-friendly innovation enhance brand equity. It highlights the distinct mediating roles of green brand image and consumer trust, emphasizing the importance of transparent communication in aligning sustainability initiatives with consumer perceptions and supporting informed strategic decision-making. Naveen Kumar R, Neerupa C, Muthukrishnan R, Vengatesan G and Muthu Kumar T -
Covid monitoring system based on sensed health parameters /
Patent Number: 202141032212, Applicant: Dr. Nimson Rio R.
A covid monitoring system (100), the system (100) comprising: a wearable (102) to be worn by a primary user, wherein the wearable (102) comprises: sensors (122a-122n) configured to sense signals representing health parameters of the primary user; a location detector (124) configured to detect real-time coordinates of the primary user; a controller (130) configured to. -
Secured automated voice controller based on security parameters /
Patent Number: 202241001260, Applicant: Jabez samuel F.
A smart home security system (700) comprising: a wearable (701) worn by a family user, wherein the wearable (701) comprises: sensors (711a-711e) configured to sense signals representing security parameters of the registered family user; a location detector (712) configured to detect real-time coordinates the family user. -
Fuzzy logic based system of intelligent electric solar dyer for fruits /
Patent Number: 201941032820, Applicant: S Vairachilai.
The present invention is related to a system of intelligent electric solar dyer for fruits using fuzzy logic based algorithm processed by at least one processor of a central processing unit of the system. -
Novel booster control system for fully automotive driverless vehicle /
Patent Number: 201941039882, Applicant: Dr. Debabrata Samanta.
The present disclosure presents a novel electronic booster control system for fully automotive driverless vehicle. The it discloses a system of vacuum booster with an automotive air compressor system which comprises a compression piston, a power transmission component and a power input part with integrated to plurality of sensors , servo controllers and a central control module of the fully automotive driverless vehicle. -
Biometrically activated self defense device for women safety /
Patent Number: 20194103621, Applicant: Dr. Debabrata Samanta.
The present invention is related to biometrically activated self-defense device for women safety. The wearable personal self-defense and altering device, comprises a biometric unit, and portable source of electrical energy, an electric discharge unit and a processing unit configured to connect a mobile computing device. -
Intelligent deep-well rescue system using ultrasonic sensors /
Patent Number: 201941048191, Applicant: Dr. Debabrata Samanta.
The present invention is related to an intelligent deep-well rescue system using ultrasonic sensors. The system for rescue a human in a narrow diameter deep-well (bore-well) comprises an ultrasonic sensor module, a grappling module, a central computing unit. The objective of the present invention is to solve the problems of the prior arts in solving issues of rescue of human being from the deep-well or bore well. -
Method for face-recognition on the basis of sketch using deep convolution neural network /
Patent Number: 201941045946, Applicant: Dr. Debabrata Samanta.
The present invention relates to method for face- recognition on the basis of sketch using deep convolution neural network. The objective of the present invention is to overcome the inadequacies of the prior art in techniques for face- recognition on the basis of sketches. -
Comparative analysis of original movies and its remakes in India /
The remake is a phenomenon both well-known and immediately recognizable but in India it is not theoretically analyzed. However, by analyzing these remakes, we can understand how these films reflect some specific cultural differences between one state and other State in India. Here researcher has taken four original and its remake films to understand the phenomena of remake. The highly intensed watching the films has helped researcher to understand the difference of films original and remakes. Researcher took one Tamil Movie and its remake in Hindi and also a Malayalam Movie and its remake in Tamil. Films are Tamil Singam to Hindi Singam and Malayalam Manichithrathazhu to Tamil Chandramukhi. -
Passive control of four storied reinforced concrete structure subjected to blast loads
In the recent past in major cities all over the world, public structures are vulnerable to blast loads caused by explosions either accidentally or intentionally. The intentional causes are not on hardened military targets but on important civilian structures, like commercial, financial and civic centers. The study of reinforced concrete Ground + 3 Storied structure subjected to blast loads have gained importance, as conventionally the reinforced newlineconcrete structures are not designed for blast loads as many of the loading codes do not mandate for the same and also due to the fact that quantifying the magnitude of the blast load is difficult to estimate. However, the structures are susceptible to damage from the explosion. To protect the life of people and to minimize the damage to the structure, it has become imperative to consider the effect of blast loads too, in addition to newlinethe conventional loads, considered as per the prevailing codes, during the analysis and design of all public buildings. The charge weights of the explosive used on the structure are 8kg, 16 kg and 24kg. The equivalent blast pressure subjected on the structure is determined, to study its corresponding effects for stand-off distances of 3000 mm and 6000mm using surface blast load. The behavior of the structure is newlinestudied by varying the parameters and verified which of these parameters can newlinebe critical to the performance of the structure. The response of the ground floor + 3 upper storied reinforced concrete skeletal structure is studied for understanding the variation of the displacements, strains and stresses for the parameters considered. Using PTC CREO 3.0 the 3D modeling of structure and structural elements were generated. HYPERMESH was used for the discretization (meshing) of structure and its elements. Static analysis and blast load analysis was carried out using ANSYS.