scholarly journals Design of Pill Filming System and Automatic Pill Classification System

2021 ◽  
Author(s):  
Juhui Lee ◽  
Soyoon Kwon ◽  
Jong Hoon Kim ◽  
Kwang Gi Kim
Keyword(s):  
Deep Learning ◽  
Classification System ◽  
Controller Design ◽  
Similarity Index ◽  
Medical Personnel ◽  
Structure Design ◽  
Classification Learning ◽  
Structure Similarity Index ◽  
Structure Similarity ◽  
Index Measure

Abstract Background Invent Pill classification system that can detect and classify into one tablet unit by deep-learning and Pill filming system that generate comprehensive and multi-dimensional data for learning.Methods Pill filming system and Pill classification system, they have two chapters consisted of structure design, model introduction, and controller design. Pill classification system's structure categorization is Input box, Linear conveyor, and Output box. In Data preprocessing, a similarity map is obtained with Structure Similarity Index Measure(SSIM). And RetinaNet is used as a pill classification learning model. Mean Accuracy Precision (mAP) is 0.9842, and we take experiment about measuring the number and accuracy of the classified pills for each experimenter's classification time. Conclusion Pill filming system and Pill classification system are expected to reduce labour losses for simple tasks. It helps medical personnel focus on significant and urgent tasks. And It can contribute to experiments about that deep-learning control the mechanical device.

IMAGE SEQUENCE PREDICTION USING ANN AND RBFNN

2013 ◽  
Vol 13 (02) ◽  
pp. 1340006 ◽  
Author(s):  
NISHCHAL K. VERMA ◽  
SHIKHA SINGH
Keyword(s):  
Neural Network ◽  
Similarity Index ◽  
Image Sequence ◽  
Image Frame ◽  
Novel Approach ◽  
Canny Edge ◽  
Mean Structure ◽  
Structure Similarity Index ◽  
Artificial Neural Network Ann ◽  
Index Measure

A novel approach to predict future image frame of an image sequence is being presented. First, a method to predict the future position of a moving object in an image sequence is discussed using artificial neural network (ANN). Second, optical flow concept is used for generating complete image frame by calculating velocity of each pixel on both axes. A separate ANN (both sigmoidal and radial basis function neural network) is modeled for each pixel's velocity and predicted velocity of each pixel is then mapped to its future values and image frames are generated. The quality evaluations of predicted images are measured by Canny edge detection-based image comparison metric (CIM) and mean structure similarity index measure (MSSIM). These proposed approaches are found to generate future images up to 10 images successfully.

scholarly journals A Novel Image Processing Methodology for X-ray Image Compression and Enhancement

2022 ◽  
pp. 1-8
Author(s):  
Calvin Omind Munna
Keyword(s):  
Image Processing ◽  
Mean Squared Error ◽  
Signal To Noise Ratio ◽  
Similarity Index ◽  
X Ray ◽  
Squared Error ◽  
Structure Similarity Index ◽  
Structure Similarity ◽  
Clinical Pictures

Currently, there a growing demand of data produced and stored in clinical domains. Therefore, for effective dealings of massive sets of data, a fusion methodology needs to be analyzed by considering the algorithmic complexities. For effective minimization of the severance of image content, hence minimizing the capacity to store and communicate data in optimal forms, image processing methodology has to be involved. In that case, in this research, two compression methodologies: lossy compression and lossless compression were utilized for the purpose of compressing images, which maintains the quality of images. Also, a number of sophisticated approaches to enhance the quality of the fused images have been applied. The methodologies have been assessed and various fusion findings have been presented. Lastly, performance parameters were obtained and evaluated with respect to sophisticated approaches. Structure Similarity Index Metric (SSIM), Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR) are the metrics, which were utilized for the sample clinical pictures. Critical analysis of the measurement parameters shows higher efficiency compared to numerous image processing methods. This research draws understanding to these approaches and enables scientists to choose effective methodologies of a particular application.

scholarly journals Franken-CT: Head and Neck MR-Based Pseudo-CT Synthesis Using Diverse Anatomical Overlapping MR-CT Scans

2021 ◽  
Vol 11 (8) ◽  
pp. 3508
Author(s):  
Pedro Miguel Martinez-Girones ◽  
Javier Vera-Olmos ◽  
Mario Gil-Correa ◽  
Ana Ramos ◽  
Lina Garcia-Cañamaque ◽  
...  
Keyword(s):  
Deep Learning ◽  
Confidence Interval ◽  
Head And Neck ◽  
Similarity Index ◽  
Ground Truth ◽  
Structural Similarity ◽  
Absolute Error ◽  
Ct Scans ◽  
Learning Methods ◽  
Index Measure

Typically, pseudo-Computerized Tomography (CT) synthesis schemes proposed in the literature rely on complete atlases acquired with the same field of view (FOV) as the input volume. However, clinical CTs are usually acquired in a reduced FOV to decrease patient ionization. In this work, we present the Franken-CT approach, showing how the use of a non-parametric atlas composed of diverse anatomical overlapping Magnetic Resonance (MR)-CT scans and deep learning methods based on the U-net architecture enable synthesizing extended head and neck pseudo-CTs. Visual inspection of the results shows the high quality of the pseudo-CT and the robustness of the method, which is able to capture the details of the bone contours despite synthesizing the resulting image from knowledge obtained from images acquired with a completely different FOV. The experimental Zero-Normalized Cross-Correlation (ZNCC) reports 0.9367 ± 0.0138 (mean ± SD) and 95% confidence interval (0.9221, 0.9512); the experimental Mean Absolute Error (MAE) reports 73.9149 ± 9.2101 HU and 95% confidence interval (66.3383, 81.4915); the Structural Similarity Index Measure (SSIM) reports 0.9943 ± 0.0009 and 95% confidence interval (0.9935, 0.9951); and the experimental Dice coefficient for bone tissue reports 0.7051 ± 0.1126 and 95% confidence interval (0.6125, 0.7977). The voxel-by-voxel correlation plot shows an excellent correlation between pseudo-CT and ground-truth CT Hounsfield Units (m = 0.87; adjusted R2 = 0.91; p < 0.001). The Bland–Altman plot shows that the average of the differences is low (−38.6471 ± 199.6100; 95% CI (−429.8827, 352.5884)). This work serves as a proof of concept to demonstrate the great potential of deep learning methods for pseudo-CT synthesis and their great potential using real clinical datasets.

scholarly journals Image Enhancement and Feature Extraction of Cytomegalovirus Image using Hierarchical Ranking Convolutional Neural Network (HR-CNN)

2019 ◽  
Vol 8 (12) ◽  
pp. 3816-3820
Keyword(s):  
Neural Network ◽  
Image Quality ◽  
Quality Assessment ◽  
Image Enhancement ◽  
Mean Square Error ◽  
Image Quality Assessment ◽  
Similarity Index ◽  
Mean Square ◽  
Structure Similarity Index ◽  
Structure Similarity

Human Cytomegalovirus is becoming a common issue around the globe , mainly it deals with the infection of the fetus in the womb. Digital image processing plays a vital role in various fields especially in the field of medicine to have a better quality of image of viruses in various forms. To have better clarity of images even in microscopic images there might be some flaws in detection of viruses because of the intensities which occur due to atmospheric lights, to overcome the flaws in microscopic images there comes a technique image enhancement to overcome noise in images especially distortion free images to be produced based on some image quality assessment and to reduce noise in an image without any loss of information. In this paper the proposed methodology called Hierarchical Ranking Convolution Neural Network is introduced based on Upward/Downward hierarchy and Forward/Backward Hierarchy to extract features and to provide intensified image of the virus. Image quality assessment is done with the parameters and evaluated using Mean Square Error, Peak signal to Noise Ratio, Root Mean Square Error, Structure Similarity Index, Mean Structure Similarity Index to prove the accuracy.

scholarly journals Hybrid Patch Propagation and Diffusion Technique for Image Inpainting

2019 ◽  
Vol 8 (3S2) ◽  
pp. 140-146
Keyword(s):  
Similarity Index ◽  
Image Inpainting ◽  
Diffusion Method ◽  
Hybridization Technique ◽  
Hybrid Technique ◽  
The Novel ◽  
Diffusion Technique ◽  
Structure Similarity Index ◽  
Structure Similarity ◽  
And Diffusion

This paper deals with anovel hybrid technique based onpatch propagation and diffusion method for image inpainting. The presented technique is used to reconstruct a damaged image. Image inpainting is a method to fill the hole with the best plausible, which is created due to damage. The novel hybridization technique of diffusion-based and exemplar-based is presented to overcome the existing problem of inpainting. The method is tested on the image dataset of TUM-IID. The performance of present method is measured using quality factor (QF) analysis, peak signal-noise ratio (PSNR), and comparing similarity by structure similarity index (SSIM). Result demonstrates that the proposed calculation performed better contrast with the existing exemplar-based technique.

scholarly journals Image Denoising via Asymptotic Nonlocal Filtering

2015 ◽  
Vol 2015 ◽  
pp. 1-9
Author(s):  
Xiaoyan Liu ◽  
Xiangchu Feng ◽  
Xuande Zhang ◽  
Xiaoping Li ◽  
Liang Luo
Keyword(s):  
Image Denoising ◽  
Similarity Index ◽  
Computation Complexity ◽  
Visual Effects ◽  
Nonlocal Means ◽  
Filtering Algorithm ◽  
Similarity Estimation ◽  
Structure Similarity Index ◽  
Structure Similarity ◽  
Influence Of Noise

The nonlocal means algorithm is widely used in image denoising, but this algorithm does not work well for high-intensity noise. To overcome this shortcoming, we establish a coupled iterative nonlocal means model in this paper. Considering the computation complexity of the new model, we realize it by using multiscale wavelet transform and propose an asymptotic nonlocal filtering algorithm which can reduce the influence of noise on similarity estimation and computation complexity. Moreover, we build a new nonlocal weight function based on the structure similarity index. Simulation results indicate that the proposed approach cannot only remove the noise but also preserve the structure of image and has good visual effects, especially for highly degenerated images.

Large Eddy Simulation of an N-Dodecane Spray Flame Under Different Ambient Oxygen Conditions

2015 ◽  
Author(s):  
Yuanjiang Pei ◽  
Bing Hu ◽  
Sibendu Som
Keyword(s):  
Large Eddy Simulation ◽  
Similarity Index ◽  
Rich Mixture ◽  
Eddy Simulation ◽  
Spray Flame ◽  
Ambient Oxygen ◽  
Large Eddy ◽  
Structure Similarity Index ◽  
Structure Similarity ◽  
Oxygen Concentrations

An n-dodecane spray flame was simulated using a dynamic structure large eddy simulation (LES) model coupled with a detailed chemistry combustion model to understand the ignition processes and the quasi-steady state flame structures. This study focuses on the effect of different ambient oxygen concentrations, 13%, 15% and 21% at an ambient temperature of 900 K and an ambient density of 22.8 kg/m3, which are typical diesel-engine relevant conditions with different levels of exhaust gas recirculation (EGR). The liquid spray was treated with a traditional Lagrangian method. A 103-species skeletal mechanism was used for the n-dodecane chemical kinetic model. It is observed that the main ignitions occur in rich mixture and the flames are thickened around 35 to 40 mm off the spray axis due to the enhanced turbulence induced by the strong recirculation upstream, just behind the head of the flames at different oxygen concentrations. At 1 ms after the start of injection, the soot production is dominated by the broader region of high temperature in rich mixture instead of the stronger oxidation of the high peak temperature. Multiple realizations were performed for the 15% O2 condition to understand the realization to realization variation and to establish best practices for ensemble-averaging diesel spray flames. Two indexes are defined. The structure-similarity index analysis suggests at least 5 realizations are needed to obtain 99% similarity for mixture fraction if the average of 16 realizations are used as the target at 0.8 ms. However, this scenario may be different for different scalars of interest. It is found that 6 realizations would be enough to reach 99% of similarity for temperature, while 8 and 14 realizations are required to achieve 99% similarity for soot and OH mass fraction, respectively. Similar findings are noticed at 1 ms. More realizations are needed for the magnitude-similarity index for the similar level of similarity as the structure-similarity index.

Sign in / Sign up

Export Citation Format

Share Document