The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algor... more The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algorithm. The distance traveled by the vehicle is calculated using the movement of the centroid over the frames and then the speed of the vehicle is estimated. The application is developed using MATLAB and SIMULINK. The average speed of a car is determined from various frames (here, considering total number of frames is 50).
International Journal for Research in Applied Science and Engineering Technology, Jun 30, 2023
This paper focuses on developing an image captioning system using deep learning techniques. The p... more This paper focuses on developing an image captioning system using deep learning techniques. The paper aims to generate descriptive textual captions for images, enabling machines to understand and communicate the content of visual data. The methodology involves leveraging convolutional neural networks (CNNs) for image feature extraction and recurrent neural networks (RNNs) for sequential language generation. The paper includes steps such as dataset collection, data preprocessing, CNN feature extraction, RNN-based captioning model implementation, model evaluation using metrics like BLEU score and METEOR, and presenting the results obtained. The expected deliverables include a functional image captioning system, comprehensive documentation, and a well-documented codebase. Through this paper, students gain practical experience in deep learning, computer vision, and natural language processing, contributing to advancements in image understanding and humanmachine interaction with visual data.
International Journal for Research in Applied Science and Engineering Technology
Diabetic Eye Diseases, such as diabetic retinopathy, pose a significant threat to the vision of i... more Diabetic Eye Diseases, such as diabetic retinopathy, pose a significant threat to the vision of individuals with diabetes. Early detection of these diseases is crucial for effective treatment and prevention of severe vision loss. In this study, we propose a novel approach for the early detection of Diabetic Eye Diseases using Local Binary Patterns (LBP) feature extraction. The LBP method captures local texture patterns in retinal images, allowing for the identification of characteristic features associated with the diseases. To evaluate the effectiveness of our approach, we utilized a dataset consisting of retinal images from diabetic patients with varying stages of the diseases. The dataset was pre-processed using standard techniques, and LBP features were extracted from the segmented retinal regions. A classification algorithm such as SVM, Random Forest, Adaboost was employed to differentiate between normal and diseased retinal images based on the extracted LBP features. Our experimental results demonstrated the efficacy of the proposed LBP feature extraction method, achieving a high accuracy of 79% using Adaboost in MESSIDOR and 85% using SVM in DIARETDB0 dataset in early detection of Diabetic Eye Diseases.
International Journal for Research in Applied Science and Engineering Technology
The most often utilized strategies for current deep learning models to accomplish a multitude of ... more The most often utilized strategies for current deep learning models to accomplish a multitude of activities on devices are mobile networks and binary neural networks. In this research, we propose a method for identifying an object using the pretrained deep learning model MobileNet for Single Shot Multi-Box Detector (SSD). This technique is utilized for real-time detection as well as webcams to detect the object in a video feed.To construct the module, we use the MobileNet and SSD frameworks to provide a faster and effective deep learning-based object detection approach.Deep learning has evolved into a powerful machine learning technology that incorporates multiple layers of features or representations of data to get cuttingedge results. Deep learning has demonstrated outstanding performance in a variety of fields, including picture classification, segmentation, and object detection. Deep learning approaches have recently made significant progress in fine-grained picture categorization, which tries to differentiate subordinate-level categories.The major goal of our study is to investigate the accuracy of an object identification method called SSD, as well as the significance of a pre-trained deep learning model called MobileNet.To perform this challenge of detecting an item in an image or video, I used OpenCV libraries, Python, and NumPy. This enhances the accuracy of behavior recognition at a processing speed required for real-time detection and daily monitoring requirements indoors and outdoors.
International Journal for Research in Applied Science and Engineering Technology
Over the last decades, the incidence of skin cancer, melanoma and non-melanoma, has increased at ... more Over the last decades, the incidence of skin cancer, melanoma and non-melanoma, has increased at a continuous rate. In particular for melanoma, the deadliest type of skin cancer, early detection is important to increase patient prognosis. Recently, deep neural networks (DNNs) have become viable to deal with skin cancer detection. In this work, we present a smartphone-based application to assist on skin cancer detection. This application is based on a Convolutional Neural Network (CNN) trained on clinical images and patients demographics, both collected from smartphones. Also, as skin cancer datasets are imbalanced, we present an approach, based on the mutation operator of Differential Evolution (DE) algorithm, to balance data. In this sense, beyond provides a flexible tool to assist doctors on skin cancer screening phase, the method obtains promising results with a balanced accuracy of 85% and a recall of 96%.
2012 NATIONAL CONFERENCE ON COMPUTING AND COMMUNICATION SYSTEMS, 2012
This paper puts forward a method to extract information from input image and further analyse usin... more This paper puts forward a method to extract information from input image and further analyse using that information. However it is relatively difficult to extract the information as the segmentation technique required, is variable between images; a limit of segmentation performance. The main objective of our paper is to propose an algorithm on how to extract that relative information out of given images efficiently and then finally use that information to identify the nature of the images. We thus propose a method of edge-based segmentation in which the image is first reconstructed, it is then converted to the L*a*b* colour space, then the Intensity transformation and the Sobel edge detection is performed and finally the information is extracted using the eight-component connected method. Using this extracted information basically the pixel content and then matching it with the information from our trained files in the database we now can identify the nature of the images.
Steganography is the process of hiding one file inside another such that others can neither ident... more Steganography is the process of hiding one file inside another such that others can neither identify the meaning of the embedded object, nor even recognize its existence. In this paper we want to increase the capacity and performance of DCT based steganography by using modified JPEG quantization table embedding technique in frequency domain. Our experimental results show that the proposed method will provide a better result in terms of some evaluation criteria named Peak Signal to noise ratio, square of error between cover & stego image, maximum no of bits that can be embedded in a cover image with a negligible probability of detection by an adversary of steganography system. Our proposed method shows that after modification of frequency values in default JPEG quantization table we can be able to increase the image quality by increasing the value of Peak signal to noise ratio, capacity and also by decreasing mean square error of embedded image. So that's why when we try to extract hidden data from stego image in receiving end we just get almost exact hidden data with less error possibilities. In our proposed method we first create modified default (8x8) quantization table from default 8X8 traditional JPEG quantization table and then by using the table we can also try to modify embedding as well as extracting algorithm in frequency domain steganography. It has been found that capacity which is the amount of information embedding in color images increases as the number of modified quantized DCT coefficients increases. So more data can be embedded in this innovative approach.
2018 Emerging Trends in Electronic Devices and Computational Techniques (EDCT), 2018
Human activity recognition plays a significant role in human-to-human interaction and interperson... more Human activity recognition plays a significant role in human-to-human interaction and interpersonal relations. Because it provides information about the identity of a person, their personality, and psychological state, it is difficult to extract. The human ability to recognize another person's activities is one of the main subjects of study of the scientific areas of computer vision and machine learning. As a result of this research, many applications, including video surveillance systems, human-computer interaction, and robotics for human behavior characterization, require a multiple activity recognition system. In image and video analysis, human activity recognition is an important research direction. In the past, a large number of papers have been published on human activity recognition in video and image sequences. In this paper, we provide a comprehensive survey of the recent development of the techniques, including methods, systems, and quantitative evaluation of the perfo...
The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algor... more The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algorithm. The distance traveled by the vehicle is calculated using the movement of the centroid over the frames and then the speed of the vehicle is estimated. The application is developed using MATLAB and SIMULINK. The average speed of a car is determined from various frames (here, considering total number of frames is 50).
International Journal for Research in Applied Science and Engineering Technology, 2017
As we all know that we are slowly moving toward automation and Automation is one of the trending ... more As we all know that we are slowly moving toward automation and Automation is one of the trending topics. So basically in this project we will be basically controlling fan speed with respect to the temperature. The system will get the temperature from the temperature Sensor and it will control the speed according to the temperature, set by the user. In this project, microcontroller forms the processing part, which firstly senses the temperature and the controller then compares the data with the set temperature. If the current temperature is greater than the set temperature, the controller turns ON the fan and the set speed will be proportional to the difference between between the set temperature & the current temperature. If the current temperature is less than the set temperature, the fan will be turned OFF .The fan's speed will change according to the temperature.
International Journal for Research in Applied Science and Engineering Technology
The communication between the robot and the android application is carried over by the Bluetooth ... more The communication between the robot and the android application is carried over by the Bluetooth link between the phone's Bluetooth and the Bluetooth device in the Robot. The ASCII commands are sent from the phone to the Robot which in turn checked by the Arduino for the control of the wheels according to the commands to move the robot in the desired direction.
In this paper, we are planning to build an application through which victims of Covid-19 can put ... more In this paper, we are planning to build an application through which victims of Covid-19 can put their data. Once an effective data set is built, the application is capable to predict a COVID-19 zone based on these data by using a regression model. The parameters of the data set will be: 1. How many times a person visited outside. 2. City of the person, who visited abroad recently if yes then which country (take the ratio of victims to their population density as a parameter), if no then we will put a zero. 3. The cases in their home town by simply accessing their location (take the ratio of victims to the population density of that area). 4. Once data set is built, then train a regression model to predict the chances of a person being a victim of Covid - 19 and once that information is available we can simply observe in which area the no of suspected no of people is greater through which we can tell the chance of an area being affected by this situation. 5. Although parameters of t...
The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algor... more The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algorithm. The distance traveled by the vehicle is calculated using the movement of the centroid over the frames and then the speed of the vehicle is estimated. The application is developed using MATLAB and SIMULINK. The average speed of a car is determined from various frames (here, considering total number of frames is 50).
International Journal for Research in Applied Science and Engineering Technology, Jun 30, 2023
This paper focuses on developing an image captioning system using deep learning techniques. The p... more This paper focuses on developing an image captioning system using deep learning techniques. The paper aims to generate descriptive textual captions for images, enabling machines to understand and communicate the content of visual data. The methodology involves leveraging convolutional neural networks (CNNs) for image feature extraction and recurrent neural networks (RNNs) for sequential language generation. The paper includes steps such as dataset collection, data preprocessing, CNN feature extraction, RNN-based captioning model implementation, model evaluation using metrics like BLEU score and METEOR, and presenting the results obtained. The expected deliverables include a functional image captioning system, comprehensive documentation, and a well-documented codebase. Through this paper, students gain practical experience in deep learning, computer vision, and natural language processing, contributing to advancements in image understanding and humanmachine interaction with visual data.
International Journal for Research in Applied Science and Engineering Technology
Diabetic Eye Diseases, such as diabetic retinopathy, pose a significant threat to the vision of i... more Diabetic Eye Diseases, such as diabetic retinopathy, pose a significant threat to the vision of individuals with diabetes. Early detection of these diseases is crucial for effective treatment and prevention of severe vision loss. In this study, we propose a novel approach for the early detection of Diabetic Eye Diseases using Local Binary Patterns (LBP) feature extraction. The LBP method captures local texture patterns in retinal images, allowing for the identification of characteristic features associated with the diseases. To evaluate the effectiveness of our approach, we utilized a dataset consisting of retinal images from diabetic patients with varying stages of the diseases. The dataset was pre-processed using standard techniques, and LBP features were extracted from the segmented retinal regions. A classification algorithm such as SVM, Random Forest, Adaboost was employed to differentiate between normal and diseased retinal images based on the extracted LBP features. Our experimental results demonstrated the efficacy of the proposed LBP feature extraction method, achieving a high accuracy of 79% using Adaboost in MESSIDOR and 85% using SVM in DIARETDB0 dataset in early detection of Diabetic Eye Diseases.
International Journal for Research in Applied Science and Engineering Technology
The most often utilized strategies for current deep learning models to accomplish a multitude of ... more The most often utilized strategies for current deep learning models to accomplish a multitude of activities on devices are mobile networks and binary neural networks. In this research, we propose a method for identifying an object using the pretrained deep learning model MobileNet for Single Shot Multi-Box Detector (SSD). This technique is utilized for real-time detection as well as webcams to detect the object in a video feed.To construct the module, we use the MobileNet and SSD frameworks to provide a faster and effective deep learning-based object detection approach.Deep learning has evolved into a powerful machine learning technology that incorporates multiple layers of features or representations of data to get cuttingedge results. Deep learning has demonstrated outstanding performance in a variety of fields, including picture classification, segmentation, and object detection. Deep learning approaches have recently made significant progress in fine-grained picture categorization, which tries to differentiate subordinate-level categories.The major goal of our study is to investigate the accuracy of an object identification method called SSD, as well as the significance of a pre-trained deep learning model called MobileNet.To perform this challenge of detecting an item in an image or video, I used OpenCV libraries, Python, and NumPy. This enhances the accuracy of behavior recognition at a processing speed required for real-time detection and daily monitoring requirements indoors and outdoors.
International Journal for Research in Applied Science and Engineering Technology
Over the last decades, the incidence of skin cancer, melanoma and non-melanoma, has increased at ... more Over the last decades, the incidence of skin cancer, melanoma and non-melanoma, has increased at a continuous rate. In particular for melanoma, the deadliest type of skin cancer, early detection is important to increase patient prognosis. Recently, deep neural networks (DNNs) have become viable to deal with skin cancer detection. In this work, we present a smartphone-based application to assist on skin cancer detection. This application is based on a Convolutional Neural Network (CNN) trained on clinical images and patients demographics, both collected from smartphones. Also, as skin cancer datasets are imbalanced, we present an approach, based on the mutation operator of Differential Evolution (DE) algorithm, to balance data. In this sense, beyond provides a flexible tool to assist doctors on skin cancer screening phase, the method obtains promising results with a balanced accuracy of 85% and a recall of 96%.
2012 NATIONAL CONFERENCE ON COMPUTING AND COMMUNICATION SYSTEMS, 2012
This paper puts forward a method to extract information from input image and further analyse usin... more This paper puts forward a method to extract information from input image and further analyse using that information. However it is relatively difficult to extract the information as the segmentation technique required, is variable between images; a limit of segmentation performance. The main objective of our paper is to propose an algorithm on how to extract that relative information out of given images efficiently and then finally use that information to identify the nature of the images. We thus propose a method of edge-based segmentation in which the image is first reconstructed, it is then converted to the L*a*b* colour space, then the Intensity transformation and the Sobel edge detection is performed and finally the information is extracted using the eight-component connected method. Using this extracted information basically the pixel content and then matching it with the information from our trained files in the database we now can identify the nature of the images.
Steganography is the process of hiding one file inside another such that others can neither ident... more Steganography is the process of hiding one file inside another such that others can neither identify the meaning of the embedded object, nor even recognize its existence. In this paper we want to increase the capacity and performance of DCT based steganography by using modified JPEG quantization table embedding technique in frequency domain. Our experimental results show that the proposed method will provide a better result in terms of some evaluation criteria named Peak Signal to noise ratio, square of error between cover & stego image, maximum no of bits that can be embedded in a cover image with a negligible probability of detection by an adversary of steganography system. Our proposed method shows that after modification of frequency values in default JPEG quantization table we can be able to increase the image quality by increasing the value of Peak signal to noise ratio, capacity and also by decreasing mean square error of embedded image. So that's why when we try to extract hidden data from stego image in receiving end we just get almost exact hidden data with less error possibilities. In our proposed method we first create modified default (8x8) quantization table from default 8X8 traditional JPEG quantization table and then by using the table we can also try to modify embedding as well as extracting algorithm in frequency domain steganography. It has been found that capacity which is the amount of information embedding in color images increases as the number of modified quantized DCT coefficients increases. So more data can be embedded in this innovative approach.
2018 Emerging Trends in Electronic Devices and Computational Techniques (EDCT), 2018
Human activity recognition plays a significant role in human-to-human interaction and interperson... more Human activity recognition plays a significant role in human-to-human interaction and interpersonal relations. Because it provides information about the identity of a person, their personality, and psychological state, it is difficult to extract. The human ability to recognize another person's activities is one of the main subjects of study of the scientific areas of computer vision and machine learning. As a result of this research, many applications, including video surveillance systems, human-computer interaction, and robotics for human behavior characterization, require a multiple activity recognition system. In image and video analysis, human activity recognition is an important research direction. In the past, a large number of papers have been published on human activity recognition in video and image sequences. In this paper, we provide a comprehensive survey of the recent development of the techniques, including methods, systems, and quantitative evaluation of the perfo...
The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algor... more The vehicle motion is detected and tracked along the frames using the modified Lucas-Kanade algorithm. The distance traveled by the vehicle is calculated using the movement of the centroid over the frames and then the speed of the vehicle is estimated. The application is developed using MATLAB and SIMULINK. The average speed of a car is determined from various frames (here, considering total number of frames is 50).
International Journal for Research in Applied Science and Engineering Technology, 2017
As we all know that we are slowly moving toward automation and Automation is one of the trending ... more As we all know that we are slowly moving toward automation and Automation is one of the trending topics. So basically in this project we will be basically controlling fan speed with respect to the temperature. The system will get the temperature from the temperature Sensor and it will control the speed according to the temperature, set by the user. In this project, microcontroller forms the processing part, which firstly senses the temperature and the controller then compares the data with the set temperature. If the current temperature is greater than the set temperature, the controller turns ON the fan and the set speed will be proportional to the difference between between the set temperature & the current temperature. If the current temperature is less than the set temperature, the fan will be turned OFF .The fan's speed will change according to the temperature.
International Journal for Research in Applied Science and Engineering Technology
The communication between the robot and the android application is carried over by the Bluetooth ... more The communication between the robot and the android application is carried over by the Bluetooth link between the phone's Bluetooth and the Bluetooth device in the Robot. The ASCII commands are sent from the phone to the Robot which in turn checked by the Arduino for the control of the wheels according to the commands to move the robot in the desired direction.
In this paper, we are planning to build an application through which victims of Covid-19 can put ... more In this paper, we are planning to build an application through which victims of Covid-19 can put their data. Once an effective data set is built, the application is capable to predict a COVID-19 zone based on these data by using a regression model. The parameters of the data set will be: 1. How many times a person visited outside. 2. City of the person, who visited abroad recently if yes then which country (take the ratio of victims to their population density as a parameter), if no then we will put a zero. 3. The cases in their home town by simply accessing their location (take the ratio of victims to the population density of that area). 4. Once data set is built, then train a regression model to predict the chances of a person being a victim of Covid - 19 and once that information is available we can simply observe in which area the no of suspected no of people is greater through which we can tell the chance of an area being affected by this situation. 5. Although parameters of t...
Uploads
Papers by Moloy Dhar