Weight Map Research Articles

Modified atmosphere packaging with chitosan coating to prevent deterioration of fresh in-hull Badami’s pistachio fruit

BackgroundFresh in-hull pistachio fruits are very perishable and hence have a limited storage life, with marketers and customers wishing for a longer shelf life. The present research was performed to explore the combined effects of different packaging conditions (ambient atmospheric condition, and passive- and active-modified atmosphere packaging) as well as chitosan coating (0, 1, and 1.5%w/v) on postharvest quality of fresh in-hull pistachios stored at 5 °C and 85–90% relative humidity for 45 days. The efficacy of treatments in prolongation of pistachio fruit shelf life was evaluated by determining weight loss, kernel firmness, acid (AV) and peroxide values (PV), mold and yeast count, aflatoxin content, and hull color parameters (L*, a*, b*, C*, h°, and BI) at 0, 15, 30 and 45 days of storage. In addition, O2 and CO2 concentrations in MAP headspace were monitored.ResultsBy applying high molecular weight chitosan coatings and MAP conditions, PV and AV values, microbial growth, aflatoxin B1 production, and weight loss values were inhibited, as well as kernel firmness reduction and hull color deterioration during storage were controlled. Active-MAP treatments in most cases performed better than passive-MAP treatments in this regard (P < 0.05). Combining chitosan coating with MAP demonstrated additive synergistic effects. At the end of storage, 1.5% chitosan-coated treatment under active-MAP maintained firmness (87% of the control treatment), hull lightness (76% of the control treatment), and microbial count (similar to the control treatment). It also maintained minimum weight loss (6.39%), acid (0.91 meq O2 kg−1 oil) and peroxide (0.49 goleic acid/100 goil) values compared to other treatments, proving that chitosan coating combined with MAP was more effective in preserving pistachio quality parameters. The equilibrium O2 and CO2 levels of passive-MAP treatments were maintained at higher values than those of active-MAP treatments which indicated that passive-MAP treatments had higher respiration rates than those of active-MAP treatments during storage.ConclusionsThe results revealed that applying the combination technique of MAP and chitosan coating effectively preserved the quality attributes and lengthened the shelf life of fresh in-hull pistachios. Therefore, it could potentially be commercialized as a new approach for future industry application.Graphical

Multimodal registration of ultrasound and MR images using weighted self-similarity structure vector

ProposeMultimodal registration of 2D Ultrasound (US) and 3D Magnetic Resonance (MR) for fusion navigation can improve the intraoperative detection accuracy of lesion. However, multimodal registration remains a challenge because of the poor US image quality. In the study, a weighted self-similarity structure vector (WSSV) is proposed to registrate multimodal images. MethodThe self-similarity structure vector utilizes the normalized distance of symmetrically located patches in the neighborhood to describe the local structure information. The texture weights are extracted using the local standard deviation to reduce the speckle interference in the US images. The multimodal similarity metric is constructed by combining a self-similarity structure vector with a texture weight map. ResultsExperiments were performed on US and MR images of the liver from 88 groups of data including 8 patients and 80 simulated samples. The average target registration error was reduced from 14.91 ± 3.86 mm to 4.95 ± 2.23 mm using the WSSV-based method. ConclusionsThe experimental results show that the WSSV-based registration method could robustly align the US and MR images of the liver. With further acceleration, the registration framework can be potentially applied in time-sensitive clinical settings, such as US-MR image registration in image-guided surgery.

Boosting deep image prior by integrating external and internal image priors

Recently, an unsupervised method [denoted as deep image prior (DIP)] that interprets images as the output of a convolutional network with randomly initialized inputs has received much attention, as the DIP can adapt its network parameters to any given noisy image without requiring training pairs, providing a great flexibility for handling those noisy images with complex structures and low degree of self-similarity. However, the DIP method still suffers from a crucial limitation that only the noisy image is used as target image, which implies that only internal prior information is used for noise removal and leads to poor denoising performance, compared with supervised denoising models. In this work, we aim to boost the DIP method by integrating external and internal image priors. Specifically, with a given noisy image, we first exploited a state-of-the-art supervised method to denoise it and obtain its corresponding denoised image called initial denoised image. Then the initial denoised image containing rich external prior information was used as extra target image together with the given noisy image within the standard DIP framework, resulting in a highly effective unsupervised recovery process. Next, to take full advantage of the uncertainty of the DIP network, the above DIP denoising routine with the different random inputs was executed multiple times to generate enough complementary denoised images (samples). Finally, with an unsupervised weight map generative network, the generated samples were fused in a pixel-wise manner. The fused image with better image quality was treated as the final denoised image. We had verified the denoising performance of our method on a large of images from several benchmark databases as well as real-world noisy images. The comparative results show that our method outperforms the original DIP method and other unsupervised methods by a large margin, and surpasses state-of-the-art supervised counterparts with comparable peak signal-to-noise rate. Source code is publicly available at https://github.com/chenxiaojun0101/BDIP.

Learning a Coordinated Network for Detail-Refinement Multiexposure Image Fusion

Nowadays, deep learning has made rapid progress in the field of multi-exposure image fusion. However, it is still challenging to extract available features while retaining texture details and color. To address this difficult issue, in this paper, we propose a coordinated learning network for detail-refinement in an end-to-end manner. Firstly, we obtain shallow feature maps from extreme over/under-exposed source images by a collaborative extraction module. Secondly, smooth attention weight maps are generated under the guidance of a self-attention module, which can draw a global connection to correlate patches in different locations. With the cooperation of the two aforementioned used modules, our proposed network can obtain a coarse fused image. Moreover, by assisting with an edge revision module, edge details of fused results are refined and noise is suppressed effectively. We conduct subjective qualitative and objective quantitative comparisons between the proposed method and twelve state-of-the-art methods on two available public datasets, respectively. The results show that our fused images significantly outperform others in visual effects and evaluation metrics. In addition, we also perform ablation experiments to verify the function and effectiveness of each module in our proposed method. The source code can be achieved at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/lok-18/LCNDR</uri> .

Efficient Binary Weight Convolutional Network Accelerator for Speech Recognition.

Speech recognition has progressed tremendously in the area of artificial intelligence (AI). However, the performance of the real-time offline Chinese speech recognition neural network accelerator for edge AI needs to be improved. This paper proposes a configurable convolutional neural network accelerator based on a lightweight speech recognition model, which can dramatically reduce hardware resource consumption while guaranteeing an acceptable error rate. For convolutional layers, the weights are binarized to reduce the number of model parameters and improve computational and storage efficiency. A multichannel shared computation (MCSC) architecture is proposed to maximize the reuse of weight and feature map data. The binary weight-sharing processing engine (PE) is designed to avoid limiting the number of multipliers. A custom instruction set is established according to the variable length of voice input to configure parameters for adapting to different network structures. Finally, the ping-pong storage method is used when the feature map is an input. We implemented this accelerator on Xilinx ZYNQ XC7Z035 under the working frequency of 150 MHz. The processing time for 2.24 s and 8 s of speech was 69.8 ms and 189.51 ms, respectively, and the convolution performance reached 35.66 GOPS/W. Compared with other computing platforms, accelerators perform better in terms of energy efficiency, power consumption and hardware resource consumption.

A dual channel decomposition and remapping fusion model for low illumination images with a wide field of view

In this paper, we investigate the problem of low illumination image enhancement under wide field of view condition. Our contribution consists of two parts: Firstly, we propose an adaptive two-parameter luminance remapping enhancement function (A2BREF) based on biological visual properties. Secondly, we propose a simple and effective noise suppression method for composite weighted map. In this paper, an image decomposition method based on the total variational energy is used to decompose the input image into structure and detail components. We apply luminance enhancement to the structure component and noise suppression to the detail component. The two components are processed in parallel without interfering with each other. First, we transform the structure component from RGB space to HSV space, and A2BREF combines the global and local luminance information of the input image to perform an adaptive luminance remapping of the V component. Then, the mapping result and the initial V component are fused with PCA in a specific order to enhance the image brightness. Secondly, the detail component contains a lot of noise and detail information of the input image. In this paper, we extract two weight maps from different stages of the detail component by means of improved adaptive Gaussian filters, and the two weight maps are combined to obtain the final weight map. After multiplying the weight maps with the initial detail components and stretching the contrast, the ideal detail components are obtained. Finally, the processed structure and detail components are combined to obtain the final result. The experimental results show that the proposed method has advantages over the contrast algorithm in terms of both subjective visual and objective metrics.

NeuroCARE: A generic neuromorphic edge computing framework for healthcare applications.

Highly accurate classification methods for multi-task biomedical signal processing are reported, including neural networks. However, reported works are computationally expensive and power-hungry. Such bottlenecks make it hard to deploy existing approaches on edge platforms such as mobile and wearable devices. Gaining motivation from the good performance and high energy-efficiency of spiking neural networks (SNNs), a generic neuromorphic framework for edge healthcare and biomedical applications are proposed and evaluated on various tasks, including electroencephalography (EEG) based epileptic seizure prediction, electrocardiography (ECG) based arrhythmia detection, and electromyography (EMG) based hand gesture recognition. This approach, NeuroCARE, uses a unique sparse spike encoder to generate spike sequences from raw biomedical signals and makes classifications using the spike-based computing engine that combines the advantages of both CNN and SNN. An adaptive weight mapping method specifically co-designed with the spike encoder can efficiently convert CNN to SNN without performance deterioration. The evaluation results show that the overall performance, including the classification accuracy, sensitivity and F1 score, achieve 92.7, 96.7, and 85.7% for seizure prediction, arrhythmia detection and hand gesture recognition, respectively. In comparison with CNN topologies, the computation complexity is reduced by over 80.7% while the energy consumption and area occupation are reduced by over 80% and over 64.8%, respectively, indicating that the proposed neuromorphic computing approach is energy and area efficient and of high precision, which paves the way for deployment at edge platforms.

UIDEF: A real-world underwater image dataset and a color-contrast complementary image enhancement framework

Numerous underwater image enhancement methods have been proposed in recent years. However, these methods are mainly evaluated using synthetic datasets with similar degradation or real-world datasets with insufficient images. A benchmark dataset containing various degraded situations and real-world underwater images is needed to evaluate the performance of these methods. In this paper, we propose a Real-world Underwater Image Dataset (UIDEF), which consists of seven categories and eleven subcategories with 9200 real-world underwater images. These categories roughly cover the multiple degradation types and different shooting perspectives of common underwater imagery. Using this dataset, we conduct a qualitative and quantitative empirical comparison of eight state-of-the-art underwater image enhancement methods to evaluate their effectiveness and robustness. Considering that these methods cannot handle both color restoration and contrast enhancement of the underwater degraded images well, we present a color-contrast complementary image enhancement framework that consists of the adaptive color perception balance and multi-scale weighted fusion. The former procedure is essential to remove the color cast of original input images, while the latter defines four attentive weight maps for making the enhanced output images present a more comfortable visual perception. Extensive experiments have validated that our framework achieves relatively satisfactory performance in most cases. In addition, we demonstrate an additional application of UIDEF in reconstructing a wider-field underwater image based on multiple images with overlapping regions. The dataset is available at https://github.com/LaibinChang/UIDEF.git.

Multi-exposure electric power monitoring image fusion method without ghosting based on exposure fusion framework and color dissimilarity feature.

To solve the ghosting artifacts problem in dynamic scene multi-scale exposure fusion, an improved multi-exposure fusion method has been proposed without ghosting based on the exposure fusion framework and the color dissimilarity feature of this study. This fusion method can be further applied to power system monitoring and unmanned aerial vehicle monitoring. In this study, first, an improved exposure fusion framework based on the camera response model was applied to preprocess the input image sequence. Second, the initial weight map was estimated by multiplying four weight items. In removing the ghosting weight term, an improved color dissimilarity feature was used to detect the object motion features in dynamic scenes. Finally, the improved pyramid model as adopted to retain detailed information about the poor exposure areas. Experimental results indicated that the proposed method improves the performance of images in terms of sharpness, detail processing, and ghosting artifacts removal and is superior to the five existing multi-exposure image fusion (MEF) methods in quality evaluation.

Learning Cross-Scale Weighted Prediction for Efficient Neural Video Compression.

Neural video codecs have demonstrated great potential in video transmission and storage applications. Existing neural hybrid video coding approaches rely on optical flow or Gaussian-scale flow for prediction, which cannot support fine-grained adaptation to diverse motion content. Towards more content-adaptive prediction, we propose a novel cross-scale prediction module that achieves more effective motion compensation. Specifically, on the one hand, we produce a reference feature pyramid as prediction sources and then transmit cross-scale flows that leverage the feature scale to control the precision of prediction. On the other hand, for the first time, a weighted prediction mechanism is introduced even if only a single reference frame is available, which can help synthesize a fine prediction result by transmitting cross-scale weight maps. In addition to the cross-scale prediction module, we further propose a multi-stage quantization strategy, which improves the rate-distortion performance with no extra computational penalty during inference. We show the encouraging performance of our efficient neural video codec (ENVC) on several benchmark datasets. In particular, the proposed ENVC can compete with the latest coding standard H.266/VVC in terms of sRGB PSNR on UVG dataset for the low-latency mode. We also analyze in detail the effectiveness of the cross-scale prediction module in handling various video content, and provide a comprehensive ablation study to analyze those important components. Test code is available at https://github.com/USTC-IMCL/ENVC.

ARBiS: A Hardware-Efficient SRAM CIM CNN Accelerator With Cyclic-Shift Weight Duplication and Parasitic-Capacitance Charge Sharing for AI Edge Application

Computing-in-memory (CIM) relieves the Von Neumann bottleneck by storing the weights of neural networks in memory arrays. However, two challenges still exist, hindering the efficient acceleration of convolutional neural networks (CNN) in artificial intelligence (AI) edge devices. Firstly, the activations for sliding window (SW) operations in CNN still bring high memory access pressure. This can be alleviated by increasing the SW parallelism, but simple array replication suffers from poor array utilization and large peripheral circuits overhead. Secondly, the partial sums from individual CIM arrays, which are usually accumulated to obtain the final sum, introduce large latency due to enormous shift-and-add operations. Moreover, high-resolution ADCs are also needed to reduce the quantization error of partial sums, further increasing the hardware costs. In this paper, a hardware-efficient CIM accelerator, ARBiS, is proposed with improved activation reusability and bit-scalable matrix-vector-multiplication (MVM) for CNN acceleration in AI edge applications. The cyclic-shift weight duplication exploits a third dimension of receptive field (RF) depth for SW weight mapping to reduce the memory accesses of activations, improving the array utilization. The parasitic-capacitance charge sharing is employed to realize high-precision analog MVM in order to reduce the ADC cost. Compared with conventional architectures, ARBiS with parallel processing of 9 SW operations achieves 56.6%~58.8% alleviation of memory access pressure. Meanwhile, ARBiS configured with 8-bit ADCs saves 92.53%~94.53% ADC energy consumption. An ARBiS accelerator is evaluated to realize a computational efficiency (CE) of 10.28 (10.43) TOPS/mm2, an energy efficiency (EE) of 91.19 (112.36) TOPS/W with 8-bit (4-bit) ADCs, achieving <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$11.4\sim 11.7\times $ </tex-math></inline-formula> ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$11.6\sim 11.8\times $ </tex-math></inline-formula> ), <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.1\sim 3.3\times $ </tex-math></inline-formula> ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.4\sim 4\times $ </tex-math></inline-formula> ) improvements over state-of-the-art works, respectively.

Prediction of individual trigeminal pain sensitivity from gray matter structure within the sensorimotor network.

To determine whether multivariate pattern regression analysis based on gray matter (GM) images constrained to the sensorimotor network could accurately predict trigeminal heat pain sensitivity in healthy individuals. Prediction of individual pain sensitivity is of clinical relevance as high pain sensitivity is associated with increased risks of postoperative pain, pain chronification, and a poor treatment response. However, as pain is a subjective experience accurate identification of such individuals can be difficult. GM structure of sensorimotor regions have been shown to vary with pain sensitivity. It is unclear whether GM structure within these regions can be used to predict pain sensitivity. In this cross-sectional study, structural magnetic resonance images and pain thresholds in response to contact heat stimulation of the left supraorbital area were obtained from 79 healthy participants. Voxel-based morphometry was used to extract segmented and normalized GM images. These were then constrained to a mask encompassing the functionally defined resting-state sensorimotor network. The masked images and pain thresholds entered a multivariate relevance vector regression analysis for quantitative prediction of the individual pain thresholds. The correspondence between predicted and actual pain thresholds was indexed by the Pearson correlation coefficient (r) and the mean squared error (MSE). The generalizability of the model was assessed by 10-fold and 5-fold cross-validation. Non-parametric permutation tests were used to estimate significance levels. Trigeminal heat pain sensitivity could be predicted from GM structure within the sensorimotor network with significant accuracy (10-fold: r=0.53, p < 0.001, MSE=10.32, p=0.001; 5-fold: r=0.46, p=0.001, MSE=10.54, p < 0.001). The resulting multivariate weight maps revealed that accurate prediction relied on multiple widespread regions within the sensorimotor network. A multivariate pattern of GM structure within the sensorimotor network could be used to make accurate predictions about trigeminal heat pain sensitivity at the individual level in healthy participants. Widespread regions within the sensorimotor network contributed to the predictive model.

TCRNet: A Trifurcated Cascaded Refinement Network for Salient Object Detection

Due to the rapid development of deep learning, salient object detection has achieved promising results in recent years. However, how to construct a powerful saliency detection network to generate saliency maps with completely highlighted salient objects and effectively suppressed background noise still remains an open and challenging problem. In this paper, we propose a novel trifurcated cascaded refinement network (TCRNet) to explore multi-level feature fusion and global information representation. Specifically, all the side-output features from the backbone network are divided into three branches (i.e., global information, deep information and shallow information) according to feature similarities, and then the global contexts, semantic knowledge and fined details respectively generated by the three branches are integrated in a cascaded refinement way. This refinement strategy aims to guide the feature learning of important regions and suppress background noises. Moreover, we propose a cross-level feature fusion (CFF) module to fuse multi-level similar features within a same branch by exploring their interrelationship, thus capturing the important semantic information and structure details. To this end, the complementarity of similar features is utilized to refine feature at each level by sharing an encoded spatial weight map. Besides, we design a context-aware feature learning (CFL) module to capture global contexts in a single layer by taking account of both single-feature representation and multi-feature interaction, thus providing important location information of salient objects. Extensive experiments on five benchmarks are conducted to demonstrate that the proposed network TCRNet performs favorably against 20 state-of-the-art salient object detection methods on five benchmark datasets, indicating the effectiveness and superiority of our method.

SoBS-X: Squeeze-Out Bit Sparsity for ReRAM-Crossbar-Based Neural Network Accelerator

Resistive random-access-memory (ReRAM) crossbar is a promising technique for deep neural network (DNN) accelerators, thanks to its in-memory and in-situ analog computing abilities for vector–matrix multiplication-and-accumulations (VMMs). However, it is challenging for crossbar architecture to exploit the sparsity in DNNs. It is inevitably complex and costly to exploit fine-grained sparsity due to the limitation of the tightly coupled crossbar structure. As a countermeasure, we develop a novel ReRAM-based DNN accelerator, named sparse-multiplication-engine (SME), based on a hardware and software co-design framework. First, we orchestrate the bit-sparse pattern to increase the density of bit-sparsity based on existing quantization methods. Such quantized weights can be nicely generated using the alternating direction method of multipliers (ADMM) optimization during the DNN fine-tuning, which can exactly enforce bit patterns in weights. Second, we propose a novel weight mapping mechanism to slice the bits of the weight across crossbars and splice the activation results in peripheral circuits. This mechanism can decouple the tightly coupled crossbar structure and cumulate the sparsity in the crossbar. Finally, a superior squeeze-out scheme empties the crossbars mapped with highly sparse nonzeros from the previous two steps. We design the SME architecture and discuss its use for other quantization methods and different ReRAM cell technologies. We further propose a workload grouping algorithm and a pipeline to achieve workload balance among crossbar-rows that concurrently execute multiply–accumulate operations to optimize the system latency. Putting all together, with the optimized model, compared with prior state-of-the-art designs, the SME shrinks the use of crossbars up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$8.7\times $ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2.1\times $ </tex-math></inline-formula> using ResNet-50 and MobileNet-v2, respectively, and achieve average <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3.1\times $ </tex-math></inline-formula> speed up with no or little accuracy loss on ImageNet.

A 1T2R1C ReRAM CIM Accelerator With Energy-Efficient Voltage Division and Capacitive Coupling for CNN Acceleration in AI Edge Applications

Resistive random-access memory (ReRAM) is widely studied in computing-in-memory (CIM) for neural network acceleration in edge devices. However, the static current of conventional 2T2R array becomes prominent with increasing computing parallelism, leading to remarkable power and area costs. To resolve this issue, a voltage-style one-transistor-two-resistor-one-capacitor ReRAM (1T2R1C) CIM accelerator using energy-efficient voltage division and capacitive coupling is proposed for convolutional neural network (CNN) acceleration in AI edge applications. The 1T2R1C cell, which is 15% smaller than the previous 2T2R cell, comprises one selection transistor, two ReRAM resistors for 1-bit weight storage, and a MOS-based capacitor. The multiply-and-accumulation (MAC) operation is realized by voltage division within a cell and capacitive coupling across different cells on a plate line. The static current during computation can be effectively reduced by the series connection of low-resistance state (LRS) and high-resistance state (HRS) ReRAM resistors within a cell and the elimination of low resistance paths in the array. A corresponding weight mapping method is also proposed to transform the multi-bit weight based on 0/1 into that on −1/1, adapting the proposed accelerator for high-precision CNN applications. By evaluation in the 28nm technology, the proposed accelerator with a 384Kb array achieves a peak energy efficiency of 400.2 TOPS/W, ~8.4X higher than previous work. The inference accuracy of CNN reaches 96.2% on the MNIST dataset.

Associating multiple vision transformer layers for fine-grained image representation

- Accurate discriminative region proposal has an important effect for fine-grained image recognition. The vision transformer (ViT) brings about a striking effect in computer vision due to its innate multi-head self-attention mechanism. However, the attention maps are gradually similar after certain layers, and since ViT used a classification token to achieve classification, it is unable to effectively select discriminative image patches for fine-grained image classification. To accurately detect discriminative regions, we propose a novel network AMTrans, which efficiently increases layers to learn diverse features and utilizes integrated raw attention maps to capture more salient features. Specifically, we employ DeepViT as backbone to solve the attention collapse issue. Then, we fuse each head attention weight within each layer to produce an attention weight map. After that, we alternatively use recurrent residual refinement blocks to promote salient feature and then utilize the semantic grouping method to propose the discriminative feature region. A lot of experiments prove that AMTrans acquires the SOTA performance on four widely used fine-grained datasets under the same settings, involving Stanford-Cars, Stanford-Dogs, CUB-200-2011, and ImageNet.

In-situ soil texture classification and physical clay content measurement based on multi-source information fusion

Soil texture is one of the most important soil characteristics that affect soil properties. Rapid acquisition of soil texture information is of great significance for accurate farmland management. Traditional soil texture analysis methods are relatively complicated and cannot meet the requirements of temporal and spatial resolution. This research introduced a self-developed vehicle-mounted in-situ soil texture detection system, which can predict the type of soil texture and the particle composition of the texture, and obtain real-time data during the measurement process without preprocessing the soil samples. The detection system is mainly composed of a conductivity measuring device, a camera, an auxiliary mechanical structure, and a control system. The soil electrical conductivity (ECa) and the texture features extracted from the surface image were input into the embedded model to realize real-time texture analysis. In order to find the best model suitable for the detection system, measurements were carried out in three test fields in Northeast and North China to compare the performance of different models applied to the detection system. The results showed that for soil texture classification, ExtraTrees performed best, with Precision, Recall, and F1 all being 0.82. For particle content of soil texture prediction, the R2 of ExtraTrees was 0.77, and RMSE and MAPE were 74.72 and 39.58. It was observed that ECa, Moment of inertia, and Entropy had larger weights in the drawn model influence weight map, and they are the main contributors to predicting soil texture. These results showed the potential of the vehicle-mounted in-situ soil texture detection system, which can provide a basis for fast, cost-effective, and efficient soil texture analysis. Keywords: soil texture, soil sensor, electrical conductivity, soil surface image DOI: 10.25165/j.ijabe.20231601.6918 Citation: Meng C, Yang W, Ren X J, Wang D, Li M Z. In-situ soil texture classification and physical clay content measurement based on multi-source information fusion. Int J Agric & Biol Eng, 2023; 16(1): 203–211.

Medical image fusion using convolutional neural network

Medical image fusion methods combine medical pictures from many morphologies to improve the accuracy and reliability of medical diagnoses, and they are becoming more significant in a variety of clinical applications. This research introduces a convolutional neural network (CNN) based medical image fusion approach to create a fused picture with good visual quality and clear structural details. To generate a weight map, the proposed technique employs a trained Siamese convolutional network to fuse the pixel activity information of source pictures. Meanwhile, the original picture is decomposed using a contrast pyramid. Source pictures are combined using distinct spatial frequency bands and a weighted fusion operator. The suggested fusion method can successfully maintain the exact structural information of source pictures and generate excellent human visual effects, according to the findings of comparison trials.

A novel fuzzy logic, gray wolf optimization and convolutional neural network based image fusion framework using discrete frame-let transform

In this paper, a novel fuzzy logic, gray wolf optimization and convolutional neural network (CNN) based infrared and visible image framework using discrete frame-let transform is proposed. A new enhancement method using fuzzy logic has been introduced to improve the contrast and visual appearance of input images. Discrete frame-let transform has been applied on both enhanced source images in order to generate more featured multi-scale coefficients. Gray wolf optimization (GWO) based fusion strategy has been introduced for merging low frequency coefficients of both enhanced source images in order to generate better target and as well as background details in the resultant fused image. Siamese convolutional neural network is applied on enhanced source images for integrating pixel activity information in order to obtain a weight map. High frequency coefficients are merged using through CNN generated weight map via discrete frame-let decomposition and local similarity based fusion strategy for generating more detailed information in the final fused image. Finally the reconstruction of resultant fused image is obtained through using inverse discrete frame-let transform. The obtained experimental results achieved through simulation clearly demonstrates that our proposed algorithm has shown superior performance in both aspects (i.e. subjective assessment and as well as objective assessment) compared to related state of art methods.

Image Generation Method Using Weight Maps for Subjective Quality Improvement in Two-Dimensional Image Synthetic Integral Three-Dimensional Display

Image Generation Method Using Weight Maps for Subjective Quality Improvement in Two-Dimensional Image Synthetic Integral Three-Dimensional Display