scholarly journals Image Reconstruction and Evaluation of Micro-surfaces for MEMS Application

2016 ◽  
Author(s):  
Mohammad Mayyas
Keyword(s):  
Image Reconstruction ◽  
Sequential Estimation ◽  
Computationally Efficient ◽  
Bending Energy ◽  
Microelectromechanical System ◽  
Microscopic Features ◽  
High Resolution Images ◽  
Windowing Technique ◽  
Processing Techniques ◽  
Traditional Image

This article develops algorithms for the characterization and the visualization of micro-scale features by using a small number of sample points, and with a goal to mitigate for the measurement shortcomings which are often destructive or time consuming. We implement the algorithms to rapidly examine the microscopic features of a Microelectromechanical System (MEMS) surface. Such images are highly dense; therefore, traditional image processing techniques might be computationally expensive. The contribution of this research include first, we develop local and global algorithm based on modified Thin Plate Spline (TPS) model to reconstruct high resolution images of the micro-surface’s topography, and its derivatives by using low resolution images. Second, we obtain a bending energy algorithm from our modified TPS model, and use it to filter out image defects. Finally, we develop a computationally efficient Windowing technique, which combines TPS and Linear Sequential Estimation (LSE), to enhance the visualization of images. The Windowing technique allows rapid image reconstruction based on the reduction of inverse problem.

Detection of Blood-Related Diseases Using Deep Neural Nets

2019 ◽  
pp. 1-15
Author(s):  
Rajithkumar B. K. ◽  
Shilpa D. R. ◽  
Uma B. V.
Keyword(s):  
Image Processing ◽  
Biomedical Engineering ◽  
Medical Diagnosis ◽  
Blood Smear ◽  
Neural Nets ◽  
Laboratory Methods ◽  
Intelligent Algorithms ◽  
Image Processing Techniques ◽  
Processing Techniques ◽  
Traditional Image

Image processing offers medical diagnosis and it overcomes the shortcomings faced by traditional laboratory methods with the help of intelligent algorithms. It is also useful for remote quality control and consultations. As machine learning is stepping into biomedical engineering, there is a huge demand for devices which are intelligent and accurate enough to target the diseases. The platelet count in a blood sample can be done by extrapolating the number of platelets counted in the blood smear. Deep neural nets use multiple layers of filtering and automated feature extraction and detection and can overcome the hurdle of devising complex algorithms to extract features for each type of disease. So, this chapter deals with the usage of deep neural networks for the image classification and platelets count. The method of using deep neural nets has increased the accuracy of detecting the disease and greater efficiency compared to traditional image processing techniques. The method can be further expanded to other forms of diseases which can be detected through blood samples.

scholarly journals Real-Time Moving Object Detection in High-Resolution Video Sensing

Sensors ◽  
2020 ◽  
Vol 20 (12) ◽  
pp. 3591 ◽  
Author(s):  
Haidi Zhu ◽  
Haoran Wei ◽  
Baoqing Li ◽  
Xiaobing Yuan ◽  
Nasser Kehtarnavaz
Keyword(s):  
High Resolution ◽  
Object Detection ◽  
Real Time ◽  
Moving Object Detection ◽  
Moving Object ◽  
Computationally Efficient ◽  
Real Time Processing ◽  
Time Processing ◽  
Video Frames ◽  
High Resolution Images

This paper addresses real-time moving object detection with high accuracy in high-resolution video frames. A previously developed framework for moving object detection is modified to enable real-time processing of high-resolution images. First, a computationally efficient method is employed, which detects moving regions on a resized image while maintaining moving regions on the original image with mapping coordinates. Second, a light backbone deep neural network in place of a more complex one is utilized. Third, the focal loss function is employed to alleviate the imbalance between positive and negative samples. The results of the extensive experimentations conducted indicate that the modified framework developed in this paper achieves a processing rate of 21 frames per second with 86.15% accuracy on the dataset SimitMovingDataset, which contains high-resolution images of the size 1920 × 1080.

scholarly journals Compressed Domain Computationally Efficient Processing Scheme for JPEG Image Filtering

2013 ◽  
Vol 8-9 ◽  
pp. 480-489 ◽  
Author(s):  
Camelia Florea ◽  
Mihaela Gordan ◽  
Bogdan Orza ◽  
Aurel Vlaicu
Keyword(s):  
Image Filtering ◽  
Input Image ◽  
Compressed Domain ◽  
Digital Cameras ◽  
Computationally Efficient ◽  
Processing Scheme ◽  
Jpeg Image ◽  
Convolution Kernels ◽  
High Resolution Images ◽  
Efficient Processing

Image filtering is one of the principal tools used in computer vision applications. Real systems store and manipulate high resolution images in compressed forms, therefore the implementation of the entire processing chain directly in the compressed domain became essential. This includes almost always linear filtering operations implemented by convolution. Linear image filtering implementation directly on the JPEG images is challenging for several reasons, including the complexity of transposing the pixel level convolution in the compressed domain, which may increase the processing time, despite avoiding the decompression. In this paper we propose a new computationally efficient solution for JPEG image filtering (as a spatial convolution between the input image and a given kernel) directly in the DCT based compressed domain. We propose that the convolution operation to be applied just on the periodical extensions of the DCT basis images, as an off-line processing, obtaining the filtered DCT basis images, which are used in data decompression. While this doesn't solve the near block boundaries filtering artefacts for large convolution kernels, for most practical cases, it provides good quality results at a very low computational complexity. These kind of implementations can run at real-time rates/ speeds and are suitable for developments of applications on digital cameras/ DSP/FPGA.

Feasible stock trajectories: a flexible and efficient sequential estimator for use in fisheries management procedures

2012 ◽  
Vol 69 (1) ◽  
pp. 161-177 ◽  
Author(s):  
Nokome Bentley ◽  
Adam D. Langley
Keyword(s):  
Fisheries Management ◽  
Probability Distributions ◽  
Sequential Estimation ◽  
Alternative Methods ◽  
Model Parameters ◽  
Computationally Efficient ◽  
Likelihood Functions ◽  
Management Procedure ◽  
Management Procedures ◽  
Stock Recruitment

We describe a sequential estimation approach designed to be used as part of a fisheries management procedure; it is computationally efficient and able to be applied to varying types, and extents, of data. The estimator maintains a pool of stock trajectories, each having a unique combination of model parameters (e.g., stock–recruitment steepness) sampled from prior probability distributions. Each year, for each trajectory, the values of variables (e.g., current biomass) are updated and tested against specified constraints. Constraints further determine the feasibility of the trajectories by defining likelihood functions for model variables, or combinations of variables, in particular years. Trajectories that fail to meet one or more of the constraints are discarded from the pool and replaced by new trajectories. Each year, stochastic forward projections of the trajectories in the pool are used to determine an optimal catch level. The flexibility and accuracy of the estimator is evaluated using the fishery for snapper, Pagrus auratus , off northern New Zealand as a case study. The sequential nature of the algorithm suggests alternative methods of presentation for understanding and explaining the fisheries estimation process. We provide recommendations for both the evaluation and operation of management procedures that employ the estimator.

Sign in / Sign up

Export Citation Format

Share Document