Enhanced Disease Detection through Image Fusion in Solanum Tuberosum L.
An Improved Technique for Enhancement of Satellite Image
Magnetic Resonance and Computer Tomography Image Fusion using Novel Weight Maps Obtained by using Median and Guided Filters
Thresholding Techniques in Computer Vision Applications
Advancement in Brain Tumour Detection using Deep Learning Technique
Identification of Volcano Hotspots by using Resilient Back Propagation (RBP) Algorithm Via Satellite Images
Data Hiding in Encrypted Compressed Videos for Privacy Information Protection
Improved Video Watermarking using Discrete Cosine Transform
Contrast Enhancement based Brain Tumour MRI Image Segmentation and Detection with Low Power Consumption
Denoising of Images by Wavelets and Contourlets using Bi-Shrink Filter
With the advancement of Artificial Intelligence and Deep Learning techniques, Neural Style transfer has become a popular methodology used to apply complex styles and effects to images which might have been an impossible task to be done manually a few years back. After using one image to style another successfully, research has been done in order to transfer styles from still images to real time video. The major problem faced in doing this has been, unlike styling still images there were inconsistent “popping”, that is inconsistent stylization from frame to frame and "temporal inconsistency" which is flickering between consecutive stylized frames. Many video style transfer models have been succeeded in improving temporal inconsistency but have failed to guarantee fast processing speed and nice perceptual style quality at the same time. In this study, we look to incorporate multiple styles to a video for a defined specific region of interest. As many models are able to transfer styles to a whole single frame but not to a specific region. For this, a convolutional neural network with the pre-trained VGG-16 model for transfer learning approach has been used. In our results we were able to transfer style to a specific region of interest of the video frame. Also, different tests were conducted to validate the model (pre-trained VGG-16) and it has been shown for which domains it is suitable.
In this technologically advancing world, we not only require security for our data but we also need advanced security for homes. In this regard, one of the important aspects is the security enhancement using the smart door unlocking system. This smart door unlocking system uses face recognition using Raspberry Pi, for the purpose of locking or unlocking the door. Face recognition is one of the most secured systems than biometrics pattern recognition technique which is used in a large spectrum of applications. In this system, camera sensor is used for capturing the faces and an image matching algorithm is used to detect the authenticated face. For image matching we are using the LBPH (Local Binary Pattern Histogram) algorithm. This algorithm converts the image from color to a grayscale image, it then divides them into pixels and they will be stored in the matrix form in the database. When a person's face is matched, the Raspberry Pi microcontroller uses the solenoid lock to unlock the door. This system provides security and also makes door unlocking effortless for elders. The proposed system is robust from hacking attacks, as our proposed system uses the Machine Learning approach.
Rapidly growing use of multimedia products in diagnosis is affected by insufficient network and computer storage, as most of these applications use images of large data size. Compression is also important for reducing the size of images, particularly at lower bit rates, and it helps to avoid the use of additional memory and bandwidth in cloud storage. Image compression is classified into two types: lossy compression and lossless compression. The lossy compression technique can be used for medical image diagnosis while maintaining decoded image quality and achieving a high compression ratio, thus increasing device efficiency and reducing network bandwidth for transmission. In this paper, we propose a novel lossy colour image compression approach based on multi transforms such as DWT, SWT, and curvelets. The proposed work focused on an image compression technique based on DWT interpolation of high frequency sub-bands, correction of high frequency sub-band estimation using SWT high frequency sub-band and curvelets and comparison of the resulting images to current lossy compression techniques. Multi transforms compressed images have a high compression rate while maintaining good image quality.
Biometrics is the physiological and behavioural characteristics of a person, such as their face, ear, finger print, thumb, voice, signature, and so on. Generally, a person's fingerprints are used for authentication, and their face is used for identification, although this is a difficult task because people use masks to cheat and commit fraud. In this paper, we investigate multimodal biometrics using two modalities, the foot and the iris, to find the best results and improve the biometric system's quality.
Image recognition has many applications today, and is being done by using various methods and technologies. Neural networks are now popularly used for various image processing applications. QCNN (Quaternion Convolutional Neural Networks) and CNN (Convolutional Neural Networks) are two methods that can be used for image recognition. In this paper, we compare performance of QCNN and CNN in colour image recognition. It shows that QCNN can easily capture the inter pixel relations producing better RGB images whereas CNN produces worst gray scale images.
Document image analysis is an important procedure in deriving adequate preservation policies for old and ancient documents. Frequently, image enhancement techniques are employed to serve the purpose of document analysis and post-processing. In due course of time, documents may have deterioration and adequate preservation policies are of immense need, and for such, enhancement procedures are employed. This paper presents image enhancement techniques, their methodology, implementation, and results. Further, image processing is a procedure that deals with analysis of image details from proposed input to desired output. Furthermore, this analysis is the observation that involves both test and processed images by virtue of processing through an enhancement technique.
