Targets Track Predicting of IR Image by CMAC

The centroid and attitude of target must be predicted in target tracking of IR image for increasing capture probability. CMAC estimator can effectually resolve conflict between operational counts and predicting precision. CMAC estimator is trained with a linear model, then the centroid and attitude are predicted. It is trained once by actual error in each frame to reduce the estimate error. CMAC has excellent predicting precision and small operational counts, it adapts to real time processing for target tracking. The experimental results show that CMAC can accurately estimate the centroid and attitude of target. It adapts to change of model and has robustness.

Enhanced Robotic Vision System Based on Deep Learning and Image Fusion

Image fusion has become one of the interesting fields that attract researchers to integrate information from different image sources.It is involved in several applications.One of the recent applications is the robotic vision.This application necessitates image enhancement of both infrared(IR)and visible images.This paper presents a Robot Human Interaction System(RHIS)based on image fusion and deep learning.The basic objective of this system is to fuse visual and IR images for efficient feature extraction from the captured images.Then,an enhancement model is carried out on the fused image to increase its quality.Several image enhancement models such as fuzzy logic,Convolutional Neural Network(CNN)and residual network(ResNet)pre-trained model are utilized on the fusion results and they are compared with each other and with the state-of-the-art works.Simulation results prove that the fuzzy logic enhancement gives the best results from the image quality perspective.Hence,the proposed system can be considered as an efficient solution for the robotic vision problem with multi-modality images.

Delineation of Mesoscale Features of Ocean on Satellite IR Image

An ICSED (Improved Cluster Shade Edge-Detection) algorithm and a series of post-processing technique are discussed for automatic delineation of mesoscale structure of the ocean on digital IR images. The popular derivative-based edge operators are shown to be too sensitive to edge fine-structure and to weak gradients. The new edge-detection algorithm is ICSED (Improved Cluster Shade Edge-detection) method and it is found to be an excel lent edge detector that exhibits the characteristic of fine-structure rejection while retaining edge sharpness. This char acteristic is highly desirable for analyzing oceanographic satellite images. A sorting technique for separating clouds or land well from ocean at both day and night is described in order to obtain high quality mesoscale features on the IR image This procedure is evaluated on an AVHRR (Advanced Very High Resolution Radiometer) image with Kuroshio. Results and analyses show that the mesoscale features can be well identified by using ICSED algorithm.

Application of a Cloud-Texture Analysis Scheme to the Cloud Cluster Structure Recognition and Rainfall Estimation in a Mesoscale Rainstorm Process

It is thought that satellite infrared （IR） images can aid the recognition of the structure of the cloud and aid the rainfall estimation. In this article, the authors explore the application of a classification method relevant to four texture features, viz. energy, entropy, inertial-quadrature and local calm, to the study of the structure of a cloud cluster displaying a typical meso-scaie structure on infrared satellite images. The classification using the IR satellite images taken during 4-5 July 2003, a time when a meso-scale torrential rainstorm was occurring over the Yangtze River basin, illustrates that the detailed structure of the cloud cluster can be obviously seen by means of the neural network classification method relevant to textural features, and the relationship between the textural energy and rainfall indicates that the structural variation of a cloud cluster can be viewed as an exhibition of the convection intensity evolvement. These facts suggest that the scheme of following a classification method relevant to textural features applied to cloud structure studies is helpful for weather analysis and forecasting.

Temperature Uniformity,Correlations and the Flow Field in Swirling Impinging Jets

Infrared thermography,velocity and impingement pressure measurements alongside numerical modelling are used in this study to resolve(heated)surface temperature distributions of turbulent swirling impinging jets for two Reynolds numbers(Re=11600 and 24600).Whilst building upon earlier discoveries for this same geometry,this paper provides three new contributions:(1)identifying the role of impingement distance(H/D)as a deciding factor in the trade-off between more efficient heat transfer(at high swirl numbers)and achieving better substrate temperature uniformity(lower gradients),(2)developing correlations to predict Nusselt number for swirling and non-swirling cooling jets,and(3)predicting the underlying mixing field in these jets and its interplay with the thermal distributions resolved.Results indicate substrate temperature uniformity varies based on H/D and swirl intensity(S)with a significant level of thermal non-uniformity occurring in near-field impingement(H/D=1)at stronger swirl(S=0.59 and 0.74).Four correlations describing the effects of S,Re,and H on the average heat transfer and stagnation heat transfer are developed and yield accuracies of 8%and 12%,respectively.Flow recirculation near the impingement surface is predicted at H/D=1 for stronger swirl jets which disappears at other substrate distances.The peak wall shear stress reduces and the flow impingement becomes radially wider at higher H/D and S.Stronger turbulence or eddy viscosity regions for non-swirling jets(S=0)are predicted in the shear layer and entrainment regions at H/D=1,but such turbulence is confined to the impingement and wall jet regions for strongly swirling flows.

Local Extreme Complete Trio Pattern for Multimedia Image Retrieval System

This paper presents a new feature descriptor, namely local extreme complete trio pattern （LECTP） for image retrieval application. The LECTP extracts complete extreme to minimal edge information in all possible directions using trio values. The LECTP integrates the local extreme sign trio patterns （LESTP） with magnitude local operator （MLOP） for image retrieval. The performance of the LECTP is tested by conducting three experiments on Corel-5 000, Corel-10 000 and MIT-VisTex color databases, respectively. The results after investigation show a significant improvement in terms of average retrieval precision （ARP） and average retrieval rate （ARR） as compared to the other state-of-the art techniques in content based image retrieval （CBIR）.

Edge preserving super-resolution infrared image reconstruction based on L1-and L2-norms

Super-resolution （SR） is a widely used tech- nology that increases image resolution using algorithmic methods. However, preserving the local edge structure and visual quality in infrared （IR） SR images is challenging because of their disadvantages, such as lack of detail, poor contrast, and blurry edges. Traditional and advanced methods maintain the quantitative measures, but they mostly fail to preserve edge and visual quality. This paper proposes an algorithm based on high frequency layer features. This algorithm focuses on the IR image edge texture in the reconstruction process. Experimental results show that the mean gradient of the IR image reconstructed by the proposed algorithm increased by 1.5, 1.4, and 1.2 times than that of the traditional algorithm based on L1- norm, L2-norm, and traditional mixed norm, respectively. The peak signal-to-noise ratio, structural similarity index, and visual effect of the reconstructed image also improved.

A two-stage restoration of distorted underwater images using compressive sensing and image registration

Imaging through a time-varying water surface exhibits severe non-rigid geometric distortions and motion blur.Theoretically,although the water surface possesses smoothness and temporal periodicity,random fluctuations are inevitable in an actual video sequence.Meanwhile,considering the distribution of information,the image structure contributes more to the restoration.In this paper,a new two-stage restoration method for distorted underwater video sequences is presented.During the first stage,salient feature points,which are selected through multiple methods,are tracked across the frames,and the motion fields at all pixels are estimated using a compressive sensing solver to remove the periodic distortions.During the second stage,the combination of a guided filter algorithm and an image registration method is applied to remove the structural-information-oriented residual distortions.Finally,the experiment results show that the method outperforms other state-of-the-art approaches in terms of the recovery effect and time.