
This journal was previously known as IEE Proceedings - Vision, Image and Signal Processing 1994-2006. ISSN 1350-245X. more..
Latest content
-
Semantic‐aware visual consistency network for fused image harmonisation
- Author(s): Huayan Yu ; Hai Huang ; Yueyan Zhu ; Aoran Chen
- + Show details - Hide details
-
p.
(1)
AbstractWith a focus on integrated sensing, communication, and computation (ISCC) systems, multiple sensor devices collect information of different objects and upload it to data processing servers for fusion. Appearance gaps in composite images caused by distinct capture conditions can degrade the visual quality and affect the accuracy of other image processing and analysis results. The authors propose a fused‐image harmonisation method that aims to eliminate appearance gaps among different objects. First, the authors modify a lightweight image harmonisation backbone and combined it with a pretrained segmentation model, in which the extracted semantic features were fed to both the encoder and decoder. Then the authors implement a semantic‐related background‐to‐foreground style transfer by leveraging spatial separation adaptive instance normalisation (SAIN). To better preserve the input semantic information, the authors design a simple and effective semantic‐aware adaptive denormalisation (SADE) module. Experimental results demonstrate that the authors’ proposed method achieves competitive performance on the iHarmony4 dataset and benefits from the harmonisation of fused images with incompatible appearance gaps.
The different instances are captured from multiple sensory units and fused in data processing servers in ISCC system. Since the disunity between collected sensors, photo environments and transmission stability may lead to some appearance gaps in the fused image. This paper proposes a novel semantic‐aware visual style consistency network for fused image harmonisation.image
-
Research on a forward‐looking scanning imaging algorithm for a high‐speed radar platform
- Author(s): Sijia Liu and Minghai Pan
- + Show details - Hide details
-
p.
(1)
AbstractThe range and azimuth information of a target can be obtained after coherent pulse accumulation of the traditional multiframe stepped‐frequency (SF) synthesis wideband echo and spectrum analysis, and high‐resolution two‐dimensional imaging of the target can be achieved. However, the accumulation of a certain number of pulses requires a long beam dwell time, which cannot meet real‐time imaging requirements for high‐speed radar moving platforms. To solve the above problems, a scanning imaging mode is proposed by combining forward‐looking imaging and scanning imaging, and a target echo signal model with the structure of scanning stepped‐frequency is constructed. The SF pulses are grouped and transmitted according to the scanning order, and the echo pulses are sorted and reorganised. After the timing compensation and range Doppler coupling compensation are completed, the target is located and projected. The proposed imaging mode can achieve high‐resolution scanning forward‐looking imaging and can basically attain an azimuth resolution of approximately 0.1° within the forward‐looking scanning range. This imaging mode has higher real‐time performance and a larger target imaging range than the traditional methods. Moreover, the simulation results showed good performance via the scanning imaging method.
In order to solve the above problems, this paper combines the forward‐looking imaging and scanning imaging of the high‐speed platform, and proposes a scanning imaging mode and constructs a target echo signal model with a scanning stepped‐frequency structure. In this mode, the problem of poor real‐time traditional 2D imaging can be optimised, making the forward‐looking imaging of high‐speed radar more flexible.image
-
Parking space number detection with multi‐branch convolution attention
- Author(s): Yifan Guo ; Jianxun Zhang ; Yuting Lin ; Jie Zhang ; Bowen Li
- + Show details - Hide details
-
p.
(1)
AbstractWith the increase of large shopping malls, there are many large parking spaces in complex environments, which increases the difficulty of finding vehicles in such environments. To upgrade the consumer's experience, some car manufacturers have proposed detecting parking space numbers in parking spaces. The detection of parking space number in parking spaces in complex environments has problems such as the diversified background of parking space numbers, tilted direction of parking space numbers, and small parking space number scale. Since no scholar has proposed a high‐performance method for such problems, a parking space number detection model based on the multi‐branch convolutional attention is presented. Firstly, using ResNet50 as the backbone network, a multi‐branch convolutional structure is proposed in the backbone network, which aims to process and fuse the feature map through three parallel branches, and enhance the network to represent ability information by convolutional attention, learn global features to selectively strengthen the features containing helpful information, and improve the ability of the model to detect the parking space number area. Secondly, a high‐level feature enhancement unit is designed to adjust the features channel by channel, obtain more spatial correlation, and reduce the loss of information in the process of feature map generation. The data results of the model on the parking space number dataset CCAG show that the precision, recall, and F‐measure are 84.8%, 84.6%, and 84.7%, respectively, which has certain advantages for parking space number detection.
The detection of parking space number in parking spaces in complex environments has problems such as the diversified background of parking space numbers, tilted direction of parking space numbers, and small parking space number scale. The authors propose a detection model for detecting the parking area.image
-
Climate change impact assessment on groundwater level changes: A study of hybrid model techniques
- Author(s): Stephen Afrifa ; Tao Zhang ; Xin Zhao ; Peter Appiahene ; Mensah Samuel Yaw
- + Show details - Hide details
-
p.
(1)
AbstractOne of the most important sources of water supply is groundwater. However, the groundwater level (GWL) is significantly impacted by the global climate change. Therefore, under these more severe climate change conditions, the accurate and simple forecast of farmland GWL is a crucial component of agricultural water management. A hybrid model (HM) of Bayesian random forest (BRF), Bayesian support vector machine (BSVM), and Bayesian artificial neural network (BANN) is built in this study. The HM is made up of a Bayesian model averaging (BMA) and three machine learning models: random forest (RF), support vector machine (SVM), and artificial neural network. These three HMs are employed to help automate logical inference and decision‐making in business intelligence for groundwater management. For this purpose, data on 8 separate climatic factors that impact GWL changes in the study area were acquired. Nine distinct farming communities' GWL change data were utilised as the dependent variables for each model fit (community data). The effectiveness of the HM techniques was assessed using the evaluation metrics of mean absolute error (MAE), coefficient of determination (R 2), mean absolute percent error (MAPE), and root mean square error (RMSE). The model fit in Suhum had the greatest performance with the highest accuracy (R 2 varied from 0.9051 to 0.9679) and the lowest error scores (RMSE ranged from 0.0653 to 0.0727, and MAE ranged from 0.0121 to 0.0541), according to the models' evaluation results. The BRF delivered the greatest results when compared to the two independent HMs, the BSVM and BANN. Future GWL and climatic variable data may be trained using the trained HM techniques to determine the effects of climate change. Farmers, businesses, and civil society organisations might benefit from continuous monitoring of GWL data and education on climate change to help control and prevent excessive deteriorations of global climate change on GWL.
In the current study, a hybrid model of Bayesian random forest, Bayesian support vector machine, and Bayesian artificial neural network is created by combining a mathematical model of Bayesian model averaging and three machine learning models of random forest, support vector machine, and artificial neural network.image
-
Sparsity‐optimised farrow structure variable fractional delay filter for wideband array
- Author(s): Wenjing Zhou ; Mingwei Shen ; Min Xu ; Guodong Han ; Yudong Zhang
- + Show details - Hide details
-
p.
(1)
AbstractIn this paper, a new sparsity‐optimised Farrow structure variable fractional delay (SFS‐VFD) filter is proposed to address the aperture effect in wideband array. Our method is based on coefficient (anti‐)symmetry and optimises the number and orders of its sub‐filters, greatly reducing the non‐zero coefficients. The established cost function is formulated as a parametric minimisation problem with multiple regularisation constraints, and solved by the modified three‐block alternating direction multiplier method (MTB‐ADMM), which is improved by introducing core variable correction items to ensure stable and fast convergence. Experimental results show that the SFS‐VFD filter reduces the complexity of the system by decreasing the use of multipliers and adders while ensuring high delay accuracy. In wideband array, the SFS‐VFD filter effectively corrects the aperture effect and achieves precise beam pointing.
In this paper, a new sparsity‐optimised Farrow structure variable fractional delay (SFS‐VFD) filter is proposed to address the aperture effect in wideband array. It uses the modified three‐block alternating direction multiplier method (MTB‐ADMM) to determine the coefficients of the filter.image
Most downloaded

Most cited
-
Parameter estimation algorithms for dynamical response signals based on the multi-innovation theory and the hierarchical principle
- Author(s): Ling Xu and Feng Ding
-
Acoustic vector sensor: reviews and future perspectives
- Author(s): Jiuwen Cao ; Jun Liu ; Jianzhong Wang ; Xiaoping Lai
-
Two-dimensional DOA estimation for L-shaped array with nested subarrays without pair matching
- Author(s): Yang-Yang Dong ; Chun-Xi Dong ; Ying-Tong Zhu ; Guo-Qing Zhao ; Song-Yang Liu
-
Image super-resolution reconstruction using the high-order derivative interpolation associated with fractional filter functions
- Author(s): Deyun Wei
-
Convolution and correlation theorems for the two-dimensional linear canonical transform and its applications
- Author(s): Qiang Feng and Bing-Zhao Li