Low-contrast infrared image real-time enhancement based on singular value nonlinear correction

Low contrast and shortage of gray levels are two main characteristics of infrared image. According to the characteristics of infrared image, a simple and practical video format MHIVF (Mitthögskolans Interlaced Video Format) and conversion technique are introduced. This kind of video format can simplify the infrared image processing. Infrared image real-time enlargement system based on FPGA is designed and the bilinear interpolation algorithm is selected for image enlargement. The experiment results show that the designed system has a lower complexity and image processing speed is obviously increased

Download Full-text

A Novel Workflow for Geosteering a Horizontal Well in a Low Resistivity Contrast Anisotropic Environment: A Case Study in Semoga Field, Indonesia

10.2118/204547-ms ◽

2021 ◽

Author(s):

Yessica Fransisca ◽

Karinka Adiandra ◽

Vinda Manurung ◽

Laila Warkhaida ◽

M. Aidil Arham ◽

...

Keyword(s):

Real Time ◽

Horizontal Well ◽

Gamma Ray ◽

Lessons Learned ◽

Electrical Anisotropy ◽

Anisotropy Ratio ◽

Well Placement ◽

Low Contrast ◽

Low Resistivity ◽

The Subject

Abstract This paper describes the combination of strategies deployed to optimize horizontal well placement in a 40 ft thick isotropic sand with very low resistivity contrast compared to an underlying anisotropic shale in Semoga field. These strategies were developed due to previously unsuccessful attempts to drill a horizontal well with multiple side-tracks that was finally drilled and completed as a high-inclined well. To maximize reservoir contact of the subject horizontal well, a new methodology on well placement was developed by applying lessons learned, taking into account the additional challenges within this well. The first approach was to conduct a thorough analysis on the previous inclined well to evaluate each formation layer’s anisotropy ratio to be used in an effective geosteering model that could better simulate the real time environment. Correct selections of geosteering tools based on comprehensive pre-well modelling was considered to ensure on-target landing section to facilitate an effective lateral section. A comprehensive geosteering pre-well model was constructed to guide real-time operations. In the subject horizontal well, landing strategy was analysed in four stages of anisotropy ratio. The lateral section strategy focused on how to cater for the expected fault and maintain the trajectory to maximize reservoir exposure. Execution of the geosteering operations resulted in 100% reservoir contact. By monitoring the behaviour of shale anisotropy ratio from resistivity measurements and gamma ray at-bit data while drilling, the subject well was precisely landed at 11.5 ft TVD below the top of target sand. In the lateral section, wellbore trajectory intersected two faults exhibiting greater associated throw compared to the seismic estimate. Resistivity geo-signal and azimuthal resistivity responses were used to maintain the wellbore attitude inside the target reservoir. In this case history well with a low resistivity contrast environment, this methodology successfully enabled efficient operations to land the well precisely at the target with minimum borehole tortuosity. This was achieved by reducing geological uncertainty due to anomalous resistivity data responding to shale electrical anisotropy. Recognition of these electromagnetic resistivity values also played an important role in identifying the overlain anisotropic shale layer, hence avoiding reservoir exit. This workflow also helped in benchmarking future horizontal well placement operations in Semoga Field. Technical Categories: Geosteering and Well Placement, Reservoir Engineering, Low resistivity Low Contrast Reservoir Evaluation, Real-Time Operations, Case Studies

Download Full-text

Real-Time Runway Detection for Infrared Aerial Image Using Synthetic Vision and an ROI Based Level Set Method

Remote Sensing ◽

10.3390/rs10101544 ◽

2018 ◽

Vol 10 (10) ◽

pp. 1544 ◽

Cited By ~ 1

Author(s):

Changjiang Liu ◽

Irene Cheng ◽

Anup Basu

Keyword(s):

Real Time ◽

Level Set Method ◽

Level Set ◽

Infrared Image ◽

Region Of Interest ◽

Aerial Image ◽

Real Time Processing ◽

Synthetic Vision ◽

Level Set Function ◽

Rough Region

We present a new method for real-time runway detection embedded in synthetic vision and an ROI (Region of Interest) based level set method. A virtual runway from synthetic vision provides a rough region of an infrared runway. A three-thresholding segmentation is proposed following Otsu’s binarization method to extract a runway subset from this region, which is used to construct an initial level set function. The virtual runway also gives a reference area of the actual runway in an infrared image, which helps us design a stopping criterion for the level set method. In order to meet the needs of real-time processing, the ROI based level set evolution framework is implemented in this paper. Experimental results show that the proposed algorithm is efficient and accurate.

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