Cloud Model-Based Method for Infrared Image Thresholding

Traditional statistical thresholding methods, directly constructing the optimal threshold criterion using the class variance, have certain versatility but lack the specificity of practical application in some cases. To select the optimal threshold for infrared image thresholding, a simple and efficient method based on cloud model is proposed. The method firstly generates the cloud models corresponding to image background and object, respectively, and defines a novel threshold dependence criterion related with the hyper-entropy of these cloud models and then determines the optimal grayscale threshold by the minimization of this criterion. It is indicated by the experiments that, compared with selected methods, using both image thresholding and target detection, the proposed method is suitable for infrared image thresholding since it performs good results and is reasonable and effective.

Download Full-text

An Infrared Image Target Segmentation Based on Improved Threshold Method

International Journal of Circuits, Systems and Signal Processing ◽

10.46300/9106.2021.15.90 ◽

2021 ◽

Vol 15 ◽

pp. 820-828

Author(s):

Manzeng Ma ◽

Dan Liu ◽

Ruirui Zhang

Keyword(s):

Image Segmentation ◽

Target Detection ◽

Infrared Imaging ◽

Infrared Image ◽

Operation Time ◽

Optimal Threshold ◽

Threshold Method ◽

Intelligent Algorithms ◽

Infrared Image Segmentation ◽

Detection And Recognition

In recent years, infrared images have been applied in more and more extensive fields and the current research of infrared image segmentation and recognition can’t satisfy the needs of practical engineering applications. The interference of various factors on infrared detectors result in the targets detected presenting the targets of low contrast, low signal-to-noise ratio (SNR) and fuzzy edges on the infrared image, thus increasing the difficulty of target detection and recognition; therefore, it is the key point to segment the target in an accurate and complete manner when it comes to infrared target detection and recognition and it has great importance and practical value to make in-depth research in this respect. Intelligent algorithms have paved a new way for infrared image segmentation. To achieve target detection, segmentation, recognition and tracking with infrared imaging infrared thermography technology mainly analyzes such features as the grayscale, location and contour information of both background and target of infrared image, segments the target from the background with the help of various tools, extracts the corresponding target features and then proceeds recognition and tracking. To seek the optimal threshold of an image can be seen as to find the optimum value of a confinement problem. As to seek the threshold requires much computation, to seek the threshold through intelligent algorithms is more accurate. This paper proposes an automatic segmentation method for infrared target image based on differential evolution (DE) algorithm and OTSU. This proposed method not only takes into consideration the grayscale information of the image, but also pays attention to the relevant information of neighborhood space to facilitate more accurate image segmentation. After determining the scope of the optimal threshold, it integrates DE’s ability of globally searching the optimal solution. This method can lower the operation time and improve the segmentation efficiency. The simulation experiment proves that this method is very effective.

Download Full-text

A Multi-Target Detection Algorithm for Infrared Image based on Retinex and LeNet5 Neural Network

10.23940/ijpe.18.11.p16.27022710 ◽

2018 ◽

Cited By ~ 1

Author(s):

Lijun Yun

Keyword(s):

Neural Network ◽

Target Detection ◽

Infrared Image ◽

Detection Algorithm

Download Full-text

What Polarimetric Weather Radars Offer to Cloud Modelers: Forward Radar Operators and Microphysical/Thermodynamic Retrievals

Atmosphere ◽

10.3390/atmos11040362 ◽

2020 ◽

Vol 11 (4) ◽

pp. 362 ◽

Cited By ~ 1

Author(s):

Alexander V. Ryzhkov ◽

Jeffrey Snyder ◽

Jacob T. Carlin ◽

Alexander Khain ◽

Mark Pinsky

Keyword(s):

Cloud Model ◽

Weather Prediction ◽

Primary Source ◽

Polarimetric Radar ◽

Microphysical Process ◽

Weather Radars ◽

Cloud Models ◽

Cloud Modeling ◽

Bin Microphysics ◽

The Impact

The utilization of polarimetric weather radars for optimizing cloud models is a next frontier of research. It is widely understood that inadequacies in microphysical parameterization schemes in numerical weather prediction (NWP) models is a primary cause of forecast uncertainties. Due to its ability to distinguish between hydrometeors with different microphysical habits and to identify “polarimetric fingerprints” of various microphysical processes, polarimetric radar emerges as a primary source of needed information. There are two approaches to leverage this information for NWP models: (1) radar microphysical and thermodynamic retrievals and (2) forward radar operators for converting the model outputs into the fields of polarimetric radar variables. In this paper, we will provide an overview of both. Polarimetric measurements can be combined with cloud models of varying complexity, including ones with bulk and spectral bin microphysics, as well as simplified Lagrangian models focused on a particular microphysical process. Combining polarimetric measurements with cloud modeling can reveal the impact of important microphysical agents such as aerosols or supercooled cloud water invisible to the radar on cloud and precipitation formation. Some pertinent results obtained from models with spectral bin microphysics, including the Hebrew University cloud model (HUCM) and 1D models of melting hail and snow coupled with the NSSL forward radar operator, are illustrated in the paper.

Download Full-text

Comparison of Ice-Phase Microphysical Parameterization Schemes Using Numerical Simulations of Tropical Convection

Journal of Applied Meteorology ◽

10.1175/1520-0450-30.7.985 ◽

1991 ◽

Vol 30 (7) ◽

pp. 985-1004 ◽

Cited By ~ 167

Author(s):

Michale McCumber ◽

Wei-Kuo Tao ◽

Joanne Simpson ◽

Richard Penc ◽

Su-Tzai Soong

Keyword(s):

Cloud Model ◽

Type Systems ◽

Two Dimensions ◽

Tropical Convection ◽

Three Dimensions ◽

Microphysical Properties ◽

Cloud Models ◽

Optimal Mix ◽

Cloud Ice ◽

Microphysical Parameterization

Abstract A numerical cloud model is used to evaluate the performance of several ice parameterizations. Results from simulations using these schemes are contrasted with each other, with an ice-free control simulation, and with observations to determine to what extent ice physics affect the realism of these results. Two different types of tropical convection are simulated. Tropical squall-type systems are simulated in two dimensions so that a large domain can be used to incorporate a complete anvil. Nonsquall-type convective lines are simulated in three dimensions owing to their smaller horizontal scale. The inclusion of ice processes enhances the agreement of the simulated convection with some features of observed convection, including the proportion of surface rainfall in the anvil region, and the intensity and structure of the radar brightband near the melting level in the anvil. In the context of our experimental design, the use of three ice classes produces better results than two ice classes or ice-free conditions, and for the tropical cumuli, the optimal mix of the bulk ice hydrometeors is cloud ice-snow-graupel. We infer from our modeling results that application of bulk ice microphysics in cloud models might be case specific, which is a significant limitation. This can have serious ramifications for microwave interpretation of cloud microphysical properties. Generalization of ice processes may require a larger number of ice categories than we have evaluated and/or the prediction of hydrometeor concentrations or particle-size spectra.

Download Full-text

Inhomogeneous Mixing in Lagrangian Cloud Models: Effects on the Production of Precipitation Embryos

Journal of the Atmospheric Sciences ◽

10.1175/jas-d-18-0087.1 ◽

2018 ◽

Vol 76 (1) ◽

pp. 113-133 ◽

Cited By ~ 9

Author(s):

Fabian Hoffmann ◽

Takanobu Yamaguchi ◽

Graham Feingold

Keyword(s):

Turbulent Mixing ◽

Cloud Model ◽

Small Scale ◽

New Approach ◽

Air Masses ◽

Lagrangian Particles ◽

Cloud Models ◽

Large Eddy ◽

Salutary Effect ◽

Linear Eddy Model

Abstract Although small-scale turbulent mixing at cloud edge has substantial effects on the microphysics of clouds, most models do not represent these processes explicitly, or parameterize them rather crudely. This study presents a first use of the linear eddy model (LEM) to represent unresolved turbulent mixing at the subgrid scale (SGS) of large-eddy simulations (LESs) with a coupled Lagrangian cloud model (LCM). The method utilizes Lagrangian particles to provide the trajectory of air masses within LES grid boxes, while the LEM is used to redistribute these air masses among the Lagrangian particles based on the local features of turbulence, allowing for the appropriate representation of inhomogeneous to homogeneous SGS mixing. The new approach has the salutary effect of mitigating spurious supersaturations. At low turbulence intensities, as found in the early stages of an idealized bubble cloud simulation, cloud-edge SGS mixing tends to be inhomogeneous and the new approach is shown to be essential for the production of raindrop embryos. At higher turbulence intensities, as found in a field of shallow cumulus, SGS mixing tends to be more homogeneous and the new approach does not significantly alter the results, indicating that a grid spacing of 20 m may be sufficient to resolve all relevant scales of inhomogeneous mixing. In both cases, droplet in-cloud residence times are important for the production of precipitation embryos in the absence of small-scale inhomogeneous mixing, either naturally due to strong turbulence or artificially as a result of coarse resolution or by not using the LEM as an SGS model.

Download Full-text

Research on Lightweight Infrared Pedestrian Detection Model Algorithm for Embedded Platform

Security and Communication Networks ◽

10.1155/2021/1549772 ◽

2021 ◽

Vol 2021 ◽

pp. 1-7

Author(s):

Zhaoli Wu ◽

Xin Wang ◽

Chao Chen

Keyword(s):

Real Time ◽

Target Detection ◽

Pedestrian Detection ◽

Infrared Image ◽

Far Infrared ◽

Detection Algorithm ◽

Model Parameters ◽

Detection Accuracy ◽

Detection Model ◽

Embedded Platform

Due to the limitation of energy consumption and power consumption, the embedded platform cannot meet the real-time requirements of the far-infrared image pedestrian detection algorithm. To solve this problem, this paper proposes a new real-time infrared pedestrian detection algorithm (RepVGG-YOLOv4, Rep-YOLO), which uses RepVGG to reconstruct the YOLOv4 backbone network, reduces the amount of model parameters and calculations, and improves the speed of target detection; using space spatial pyramid pooling (SPP) obtains different receptive field information to improve the accuracy of model detection; using the channel pruning compression method reduces redundant parameters, model size, and computational complexity. The experimental results show that compared with the YOLOv4 target detection algorithm, the Rep-YOLO algorithm reduces the model volume by 90%, the floating-point calculation is reduced by 93.4%, the reasoning speed is increased by 4 times, and the model detection accuracy after compression reaches 93.25%.

Download Full-text

VISUALIZATION OF POINT CLOUD MODELS IN MOBILE AUGMENTED REALITY USING CONTINUOUS LEVEL OF DETAIL METHOD

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xliv-4-w1-2020-167-2020 ◽

2020 ◽

Vol XLIV-4/W1-2020 ◽

pp. 167-170

Author(s):

L. Zhang ◽

P. van Oosterom ◽

H. Liu

Keyword(s):

Augmented Reality ◽

Point Cloud ◽

Cloud Model ◽

Point Clouds ◽

Level Of Detail ◽

Mobile Augmented Reality ◽

Cloud Models ◽

Point Cloud Model ◽

Visual Artifacts ◽

Mobile Ar

Abstract. Point clouds have become one of the most popular sources of data in geospatial fields due to their availability and flexibility. However, because of the large amount of data and the limited resources of mobile devices, the use of point clouds in mobile Augmented Reality applications is still quite limited. Many current mobile AR applications of point clouds lack fluent interactions with users. In our paper, a cLoD (continuous level-of-detail) method is introduced to filter the number of points to be rendered considerably, together with an adaptive point size rendering strategy, thus improve the rendering performance and remove visual artifacts of mobile AR point cloud applications. Our method uses a cLoD model that has an ideal distribution over LoDs, with which can remove unnecessary points without sudden changes in density as present in the commonly used discrete level-of-detail approaches. Besides, camera position, orientation and distance from the camera to point cloud model is taken into consideration as well. With our method, good interactive visualization of point clouds can be realized in the mobile AR environment, with both nice visual quality and proper resource consumption.

Download Full-text