Traditional Matching Research Articles

High-Precision Disparity Estimation for Lunar Scene Using Optimized Census Transform and Superpixel Refinement

High-precision lunar scene 3D data are essential for lunar exploration and the construction of scientific research stations. Currently, most existing data from orbital imagery offers resolutions up to 0.5–2 m, which is inadequate for tasks requiring centimeter-level precision. To overcome this, our research focuses on using in situ stereo vision systems for finer 3D reconstructions directly from the lunar surface. However, the scarcity and homogeneity of available lunar surface stereo datasets, combined with the Moon’s unique conditions—such as variable lighting from low albedo, sparse surface textures, and extensive shadow occlusions—pose significant challenges to the effectiveness of traditional stereo matching techniques. To address the dataset gap, we propose a method using Unreal Engine 4 (UE4) for high-fidelity physical simulation of lunar surface scenes, generating high-resolution images under realistic and challenging conditions. Additionally, we propose an optimized cost calculation method based on Census transform and color intensity fusion, along with a multi-level super-pixel disparity optimization, to improve matching accuracy under harsh lunar conditions. Experimental results demonstrate that the proposed method exhibits exceptional robustness and accuracy in our soon-to-be-released multi-scene lunar dataset, effectively addressing issues related to special lighting conditions, weak textures, and shadow occlusion, ultimately enhancing disparity estimation accuracy.

Effective Hard Negative Mining for Contrastive Learning-based Code Search

Background. Code search aims to find the most relevant code snippet in a large codebase based on a given natural language query. An accurate code search engine can increase code reuse and improve programming efficiency. The focus of code search is how to represent the semantic similarity of code and query. With the development of code pre-trained models, the pattern of using numeric feature vectors (embeddings) to represent code semantics and using vector distance to represent semantic similarity has replaced traditional string matching methods. The quality of semantic representations is critical to the effectiveness of downstream tasks such as code search. Currently, the state-of-the-art (SOTA) learning method uses the contrastive learning paradigm. The objective of contrastive learning is to maximize the similarity between matching code and query (positive samples) and minimize the similarity between mismatched pairs (negative samples). To increase the reusing of negative samples, prior contrastive learning approaches use a large queue (memory bank) to store embeddings. Problem. However, there is still a lot of room for improvement in using negative examples for code search: ① Due to the random selection of negative samples, semantic representations learned by existing models cannot distinguish similar codes well. ② Since semantic vectors in the memory bank are reused from previous inference results and then directly used for loss function calculation without gradient descent, the model cannot effectively learn the negative sample semantic information. Method. To solve the above problems, we propose a contrastive learning code search model with hard negative mining called CoCoHaNeRe: ❶ To enable the model to distinguish similar codes, we introduce hard negative examples into contrastive training, which are negative examples in the codebase that are most similar to positive examples. As a result, hard negative examples are most likely to make the model make mistakes. ❷ To improve the learning efficiency of negative samples during training, we add all hard negative examples to the model's gradient descent process. Result. To verify the effectiveness of CoCoHaNeRe, we conducted experiments on large code search datasets with six programming languages, as well as similar retrieval tasks code clone detection and code question answering. Experimental results show that our model achieves SOTA performance. In the code search task, the average MRR score of CoCoHaNeRe exceeds CodeBERT, GraphCodeBERT, and UniXcoder by 11.25%, 8.13%, and 7.38%, respectively. It has also made great progress in code clone detection and code question answering. In addition, our method performs well in different programming languages and code pre-training models. Furthermore, qualitative analysis shows that our model effectively distinguishes high-order semantic differences between similar codes.

Automated painting color matching technology based on semantic intelligence understanding

Painting color matching technology is widely used in the production and printing process of products. Traditional painting and color matching have been unable to meet market demands. Based on this, a large-scale corpus under the existing semantic intelligent understanding system is used as the knowledge source. The computer automated painting color matching model is constructed. It is applied in case studies to address issues such as unclear query intentions, mismatched system retrieval terms, and return errors caused by uncertain factors such as synonyms and polysemy. This provides new ideas for the application of semantic intelligence understanding and automated painting color matching technology. The experimental results showed that the precision, recall, and F1 of the method used in the research were 0.8639, 0.8026, and 0.8309, respectively, significantly superior to commonly used methods. This indicates that the proposed automated painting color matching technology based on semantic intelligent understanding has high performance, which can effectively meet the painting color matching requirements.

Machine learning and propensity score matching for evaluating the effect of special education services on childrens later math performances

The goal of the research is to analyze the effect of special education service brought to childrens later math performance. The dataset is derived from Early Childhood Longitudinal Studies(ECLS) program and the children being studied come from diverse socioeconomic and racial/ethnic backgrounds. We applied PSM and modern machine learning methods including OLS, KNN, BART and MLP on the data in order to calculate the average treatment effect of special education services. It turned out that all the listed machine learning algorithms with comparatively low STD outperformed the traditional propensity score matching. KNN, BART and OLS excelled, offering much more stable calculations. The value of ATE computed through all the methods appeared below zero. By applying linear regression and PCA on all the influencing factors, the analysis revealed that the differences of some factors between the controlled and exposed groups led better math performances to appear usually in the absence of special treatment. Thus, special treatment effect led the trained model to predict lower scores, which finally caused the difference to be negative.

Intelligent Dance Motion Evaluation: An Evaluation Method Based on Keyframe Acquisition According to Musical Beat Features.

Motion perception is crucial in competitive sports like dance, basketball, and diving. However, evaluations in these sports heavily rely on professionals, posing two main challenges: subjective assessments are uncertain and can be influenced by experience, making it hard to guarantee timeliness and accuracy, and increasing labor costs with multi-expert voting. While video analysis methods have alleviated some pressure, challenges remain in extracting key points/frames from videos and constructing a suitable, quantifiable evaluation method that aligns with the static-dynamic nature of movements for accurate assessment. Therefore, this study proposes an innovative intelligent evaluation method aimed at enhancing the accuracy and processing speed of complex video analysis tasks. Firstly, by constructing a keyframe extraction method based on musical beat detection, coupled with prior knowledge, the beat detection is optimized through a perceptually weighted window to accurately extract keyframes that are highly correlated with dance movement changes. Secondly, OpenPose is employed to detect human joint points in the keyframes, quantifying human movements into a series of numerically expressed nodes and their relationships (i.e., pose descriptions). Combined with the positions of keyframes in the time sequence, a standard pose description sequence is formed, serving as the foundational data for subsequent quantitative evaluations. Lastly, an Action Sequence Evaluation method (ASCS) is established based on all action features within a single action frame to precisely assess the overall performance of individual actions. Furthermore, drawing inspiration from the Rouge-L evaluation method in natural language processing, a Similarity Measure Approach based on Contextual Relationships (SMACR) is constructed, focusing on evaluating the coherence of actions. By integrating ASCS and SMACR, a comprehensive evaluation of dancers is conducted from both the static and dynamic dimensions. During the method validation phase, the research team judiciously selected 12 representative samples from the popular dance game Just Dance, meticulously classifying them according to the complexity of dance moves and physical exertion levels. The experimental results demonstrate the outstanding performance of the constructed automated evaluation method. Specifically, this method not only achieves the precise assessments of dance movements at the individual keyframe level but also significantly enhances the evaluation of action coherence and completeness through the innovative SMACR. Across all 12 test samples, the method accurately selects 2 to 5 keyframes per second from the videos, reducing the computational load to 4.1-10.3% compared to traditional full-frame matching methods, while the overall evaluation accuracy only slightly decreases by 3%, fully demonstrating the method's combination of efficiency and precision. Through precise musical beat alignment, efficient keyframe extraction, and the introduction of intelligent dance motion analysis technology, this study significantly improves upon the subjectivity and inefficiency of traditional manual evaluations, enhancing the scientificity and accuracy of assessments. It provides robust tool support for fields such as dance education and competition evaluations, showcasing broad application prospects.

Biometric holographic encryption and authentication with multiple optics-biology keys based on inhomogeneous media optics

Biometric authentication has been widely used because of its uniqueness, accuracy, and versatility, but existing methods still have some shortcomings. Because biometric data is stored in the database, once the database is invaded, it will cause a large number of biometric data leakage and illegal use of huge risks; The cost of traditional feature extraction and matching strategy is too high; Biometric is only used for authentication, and its single function cannot meet the increasing security requirements; Methods that rely solely on biometric keys are not secure enough. To overcome these shortcomings, we propose the biometric holographic encryption and authentication with multiple optics-biology keys based on inhomogeneous media optics. The proposed method innovatively combines biometrics and inhomogeneous media optics to achieve dual functions of encryption and authentication. Instead of storing biometric data directly, the encryption and authentication information is encrypted as computer generated holograms in this method and stored in the database to reduce the risk of biometric data leakage. This method has low cost without complicated feature extraction and comparison authentication process. The dual functions of authentication and encryption is more practical. Multiple optics-biology keys improve the security. The security and robustness are analyzed. A graphical user interface (GUI) is designed for the method to promote its application. This work provides new methods and new ideas for the field of optics encryption and authentication, and will certainly have a broad application prospect.

Revolutionizing SIEM Security: An Innovative Correlation Engine Design for Multi-Layered Attack Detection.

Advances in connectivity, communication, computation, and algorithms are driving a revolution that will bring economic and social benefits through smart technologies of the Industry 4.0 era. At the same time, attackers are targeting this expanded cyberspace to exploit it. Therefore, many cyberattacks are reported each year at an increasing rate. Traditional security devices such as firewalls, intrusion detection systems (IDSs), intrusion prevention systems (IPSs), anti-viruses, and the like, often cannot detect sophisticated cyberattacks. The security information and event management (SIEM) system has proven to be a very effective security tool for detecting and mitigating such cyberattacks. A SIEM system provides a holistic view of the security status of a corporate network by analyzing log data from various network devices. The correlation engine is the most important module of the SIEM system. In this study, we propose the optimized correlator (OC), a novel correlation engine that replaces the traditional regex matching sub-module with a novel high-performance multiple regex matching library called "Hyperscan" for parallel log data scanning to improve the performance of the SIEM system. Log files of 102 MB, 256 MB, 512 MB, and 1024 MB, generated from log data received from various devices in the network, are input into the OC and simple event correlator (SEC) for applying correlation rules. The results indicate that OC is 21 times faster than SEC in real-time response and 2.5 times more efficient in execution time. Furthermore, OC can detect multi-layered attacks successfully.

Mechanism of the Terahertz Wave-MXene Interaction and Surface/Interface Chemistry of MXene for Terahertz Absorption and Shielding.

ConspectusOver the past two decades, terahertz (THz) technology has undergone rapid development, driven by advancements and the growing demand for THz applications across various scientific and technological domains. As the cornerstone of THz technology, strong THz-matter interactions, especially realized as high THz intrinsic absorption in nanometer-thick materials, play a highly important role in various applications including but not limited to THz absorption/shielding, detection, etc. The rigorous electromagnetic theory has posited a maximum intrinsic absorption of 50% for electromagnetic waves by thin films, and the succinct impedance matching condition has also been formulated to guide the design of highly intrinsically absorbing materials. However, these theories face challenges when applied to the THz spectrum with an ultrabroad bandwidth. Existing thin films typically achieve a maximum intrinsic absorption within a narrow frequency range, significantly limiting the performance of THz absorbers and detectors. To date, both theoretical frameworks and experimental solutions are lacking in overcoming the challenge of achieving broadband maximum intrinsic absorption in the THz regime.In this Account, we describe how two-dimensional (2D) transition-metal carbide and/or nitride (MXene) films with nanometer thickness can realize the maximum intrinsic absorption in the ultrabroad THz band, which successfully addresses the forementioned longstanding issue. Surprisingly, traditional DC impedance matching theory fails to explain this phenomenon, while we instead propose a novel theory of AC impedance matching to provide a satisfactory explanation. By delving into the microscopic transport behavior of free electrons in MXene, we discover that intraflake transport dominates terahertz conductivity under THz wave excitation, while interflake transport primarily dictates DC conductivity. This not only elucidates the significant disparities between DC and AC impedance in MXenes but also underscores the suitability of AC impedance matching for achieving broadband THz absorption limits. Furthermore, we identify a high electron concentration and short relaxation time as crucial factors for achieving broadband maximum absorption in the THz regime. Although approaching the THz intrinsic absorbing limits, it still faces hurdles to the use of MXene in practical applications. First, diverse and uncontrollable terminations exist on the surface of MXene stemming from the synthesis process, which largely influence the electron structure and THz absorbing property of MXene. Second, MXene suffers from poor stability in the presence of oxygen and water. To address the above issues, we have undertaken distinctive works to precisely control the terminations and suppress the oxidation of MXene even at high temperature through surface and interface chemistry, such as low-temperature Lewis basic halide treatment and building a Ti3C2Tx/extracted bentonite (EB) interface. For practical application consideration, we proposed a copolymer-polyacrylic latex (PAL)-based MXene waterborne paint (MWP) with a strong intermolecular polar interaction between MWP and the substrate provided by the cyano group in PAL. This not only has strong THz EMI shielding/absorption efficiency but also can easily adhere to various substrates that are commonly used in the THz band. These studies may have significant implications for future applications of MXene nanofilms in THz optoelectronic devices.

Evaluation of GLOCK 9 mm Firing Pin Aperture Shear Mark Individuality Based on 3,156 Different Pistols (Manufactured Over a 30 Year Period in Two Countries) Using Additional Pattern Matching and IBIS Pattern Recognition

Over a period of 30 years, a number of fired GLOCK cartridge cases have been evaluated. A total of 3156 GLOCK firearms were used to generate a sample of the same size. Our research hypothesis was that no cartridge cases fired from different 9 mm semiautomatic GLOCK pistols would be mistaken as coming from the same gun (a false match). Using optical comparison microscopy, two separate experiments were carried out to test this hypothesis. A subsample of 617 test-fired cartridge cases were subjected to algorithmic comparison by the Integrated Ballistics Identification System (IBIS). The second experiment subjected the full set of 3,156 cases to manual comparisons using traditional pattern matching. None of the cartridge cases were “matched” by either of these two experiments. Using these empirical findings, an established conservative Bayesian probability model was used to estimate the chance that a 9 mm cartridge case, fired from a GLOCK, could be mistaken as coming from the same firearm when in fact it did not (i.e., the false match probability).

Research on a Matching Method for Vehicle-Borne Laser Point Cloud and Panoramic Images Based on Occlusion Removal

Vehicle-borne mobile mapping systems (MMSs) have been proven as an efficient means of photogrammetry and remote sensing, as they simultaneously acquire panoramic images, point clouds, and positional information along the collection route from a ground-based perspective. Obtaining accurate matching results between point clouds and images is a key issue in data application from vehicle-borne MMSs. Traditional matching methods, such as point cloud projection, depth map generation, and point cloud coloring, are significantly affected by the processing methods of point clouds and matching logic. In this study, we propose a method for generating matching relationships based on panoramic images, utilizing the raw point cloud map, a series of trajectory points, and the corresponding panoramic images acquired using a vehicle-borne MMS as input data. Through a point-cloud-processing workflow, irrelevant points in the point cloud map are removed, and the point cloud scenes corresponding to the trajectory points are extracted. A collinear model based on spherical projection is employed during the matching process to project the point cloud scenes to the panoramic images. An algorithm for vectorial angle selection is also designed to address filtering out the occluded point cloud projections during the matching process, generating a series of matching results between point clouds and panoramic images corresponding to the trajectory points. Experimental verification indicates that the method generates matching results with an average pixel error of approximately 2.82 pixels, and an average positional error of approximately 4 cm, thus demonstrating efficient processing. This method is suitable for the data fusion of panoramic images and point clouds acquired using vehicle-borne MMSs in road scenes, provides support for various algorithms based on visual features, and has promising applications in fields such as navigation, positioning, surveying, and mapping.

Iterative hybrid compressive sensing-based channel estimation method for intelligent reflecting surface-supported millimeter wave systems

The Intelligent reflecting surfaces (IRSs) have been established to show the capability to enhance spectral and energy efficiency by using the passive beamforming at the IRS point and joint optimization of the active beamforming at the base station (BS). However, in the smart wireless environments in which mostly passive RISs are deplored the estimation of the channel state estimation is challenging. This paper, therefore, aims to propose a channel estimation scheme with an improved performance in comparison with some other recently proposed estimation schemes. The proposed channel estimation scheme employs the hybrid strategy to combine the improved versions of traditional Orthogonal Matching Pursuit (OMP) and Subspace Pursuit (SP) compressive sensing algorithms as a basis for the proposed estimator. The proposed estimator, through computer simulations, shows improved performance when compared with the other four channel estimators that were recently documented in literature though with a slightly high computational complexity cost.

Experimental Study for the Matching of Explosives and Rocks Based on Rock Hydrophysical Properties

The study of the hydrophysical properties of rocks is indispensable for the development of hydraulic engineering, especially for blasting operations in water. Reasonable matching between explosives and rocks increases the utilization of explosive energy and improves the blasting performances. Based on the energy law in the rock blasting process, the matching relationship between explosives and rock is studied by combining experimental and theoretical methods for the hydrophysical properties of the rock itself. Firstly, the theoretical solutions for crushing-zone energy, fragmentation energy and fragment-throwing energy are derived. Subsequently, concrete blocks are prepared with four types of cement–sand ratios, and four types of emulsion explosives are used to carry out single-hole blasting tests in which a high-speed camera is used to capture the trajectory of the blasting fragments that are later collected. Finally, the crushing energy, fracturing energy and fragment-throwing energy are calculated according to the test results and the basic parameters of the used explosives and concrete models. The results show that the size and distribution pattern of blasting blocks are significantly affected by the hydrophysical properties of concrete and explosive properties; the higher the energy consumption in the rupture zone, the smaller the size of the fragments and the more uniform the distribution. Moreover, the median utilization efficiency of explosive energy on rock breaking is 26.4%, the energy consumption in the crushing zone is approximately 8.4%, that in the rupture zone is approximately 10.9%, and that in the throwing energy of fragments accounts for approximately 7.1%. It is also found that the traditional wave impedance matching theory fails to obtain the best explosive energy utilization. On the contrary, the concrete specimen had the best fracturing effect and the highest energy utilization of 30.77% when the impedance ratio of concrete to explosives is 1.479.

CFNAM-PG: Bridging Phonetic and Glyphic Information for Chinese Full Name and Abbreviation Matching Based on Simbert and DenseNet

Matching abbreviated names with their full names (full-abbr matching) plays a key role in data integration, address matching, information retrieval, and other fields. Traditional full-abbr matching technology often encounters issues related to near homophones and near homoglyphs. First, a near-homophone full-abbr matching model based on Simbert and VGG was first proposed, which integrates character and speech features, leveraging a speech recognition model and combining a brain-like cognitive learning dual-process mechanism which involves linguistic knowledge and neural network together. Second, to address the problem of near-homoglyph full-abbr matching in Chinese, a DenseNet-based model that fuses glyph structure and image features was proposed, in which statistical feature extractors are employed to extract feature vectors for glyphic features including stroke, Wubi and structural features separately. Lastly, the near-homophone model and the near-homoglyph model are coupled to work together in the full-abbr matching task, in which expert knowledge is used as a component of the feature optimizer. Experimental results showed that the integrated model significantly increased the matching accuracy to 87.5%, demonstrating a 12.3% improvement.

An Improvement Method for Improving the Surface Defect Detection of Industrial Products Based on Contour Matching Algorithms.

Aiming at the problems of the poor robustness and universality of traditional contour matching algorithms in engineering applications, a method for improving the surface defect detection of industrial products based on contour matching algorithms is detailed in this paper. Based on the image pyramid optimization method, a three-level matching method is designed, which can quickly obtain the candidate pose of the target contour at the top of the image pyramid, combining the integral graph and the integration graph acceleration strategy based on weak classification. It can quickly obtain the rough positioning and rough angle of the target contour, which greatly improves the performance of the algorithm. In addition, to solve the problem that a large number of duplicate candidate points will be generated when the target candidate points are expanded, a method to obtain the optimal candidate points in the neighborhood of the target candidate points is designed, which can guarantee the matching accuracy and greatly reduce the calculation amount. In order to verify the effectiveness of the algorithm, functional test experiments were designed for template building function and contour matching function, including uniform illumination condition, nonlinear condition and contour matching detection under different conditions. The results show that: (1) Under uniform illumination conditions, the detection accuracy can be maintained at about 93%. (2) Under nonlinear illumination conditions, the detection accuracy can be maintained at about 91.84%. (3) When there is an external interference source, there will be a false detection or no detection, and the overall defect detection rate remains above 94%. It is verified that the proposed method can meet the application requirements of common defect detection, and has good robustness and meets the expected functional requirements of the algorithm, providing a strong technical guarantee and data support for the design of embedded image sensors in the later stage.

Automatic depth matching method of well log based on deep reinforcement learning

In the traditional well log depth matching tasks, manual adjustments are required, which means significantly labor-intensive for multiple wells, leading to low work efficiency. This paper introduces a multi-agent deep reinforcement learning (MARL) method to automate the depth matching of multi-well logs. This method defines multiple top-down dual sliding windows based on the convolutional neural network (CNN) to extract and capture similar feature sequences on well logs, and it establishes an interaction mechanism between agents and the environment to control the depth matching process. Specifically, the agent selects an action to translate or scale the feature sequence based on the double deep Q-network (DDQN). Through the feedback of the reward signal, it evaluates the effectiveness of each action, aiming to obtain the optimal strategy and improve the accuracy of the matching task. Our experiments show that MARL can automatically perform depth matches for well-logs in multiple wells, and reduce manual intervention. In the application to the oil field, a comparative analysis of dynamic time warping (DTW), deep Q-learning network (DQN), and DDQN methods revealed that the DDQN algorithm, with its dual-network evaluation mechanism, significantly improves performance by identifying and aligning more details in the well log feature sequences, thus achieving higher depth matching accuracy.

Precision oncology and AI: The ACMA system for personalized treatment matching using real-world data.

e23134 Background: Precision oncology promises tailored cancer treatments, yet current methods often overlook individual patient variability. The AI-powered Case Matching Assistant (ACMA) system addresses this by utilizing real-world data (RWD) from over 100,000 clinical cases to personalize treatment strategies, considering genetic profiles, cancer types, treatment stages, and drug interactions. Methods: ACMA enhances the Large Language Model (LLM) with oncology-specific pretraining that encompasses both molecular and clinical data, ensuring a holistic understanding of patient profiles. It employs cosine similarity for multi-faceted case comparisons, integrating genetic, biomarker, and vital clinical information. Context-aware learning prompts fine-tune the LLM, generating recommendations that reflect the full spectrum of patient-specific data, from genetic variations to clinical histories and treatment responses, facilitating truly personalized treatment pathways. Results: Preliminary results show ACMA's tailored recommendations improve patient outcomes, with increased median survival rates and cost-effective treatment options. The system's 90% relevance match for clinical trial enrollment also expands access to novel therapies. In the absence of a standard treatment protocol for BRCA germline mutations in advanced lung cancer, ACMA provides crucial insights. For instance, a 36-year-old female with BRCA1 c.527del mutation achieved complete remission with a PFS of 38 months after a combination therapy trial. A 56-year-old male with a BRCA2 S1722Yfs*mutation saw partial remission and over 6 months PFS with olaparib. Another male patient, a smoker with concurrent BRCA2 G1712X and TP53 mutations, experienced partial remission following apatinib therapy. ACMA analyzed treatment data from similar genetic profiles and identified a targeted therapy that had shown efficacy in comparable cases. The oncologist used this information as a reference to adjust the treatment plan, resulting in a positive clinical outcome for the patient. This example underscores the potential of ACMA to provide actionable insights for personalized cancer care. Conclusions: Surpassing traditional single-feature matching, ACMA rapidly aligns multi-dimensional characteristics such as cancer type, disease stage, molecular data, and treatment phases. It customizes the interplay of these features with medical logic, delivering tailored outputs through the LLM. This approach offers an economical, convenient, and timely solution, setting a new benchmark in personalized oncology and establishing ACMA as a leader in the field. This distinctive advantage not only enhances patient outcomes but also establishes a formidable barrier to entry for competitors, solidifying ACMA's position at the forefront of oncology practice.

An efficient deep learning-based workflow for real-time CO2 plume visualization in saline aquifer using distributed pressure and temperature measurements

Underground carbon dioxide (CO2) sequestration is widely accepted as a proven and established technology to respond to global warming from greenhouse gas emissions. It is crucial to monitor the CO2 plume effectively throughout the life cycle of a geologic CO2 sequestration project to ensure safety and storage efficacy. We propose a fast and efficient deep learning workflow for near real-time data assimilation, forecasting and visualization of CO2 plume evolution in saline aquifer and demonstrate its application at a field site.Our proposed workflow integrates field measurements, including distributed pressure and temperature at wells to visualize spatial and temporal migration of the CO2 plume in the subsurface. The deep learning framework has two key concepts that enhance the efficiency of the training and robustness of the CO2 plume prediction: first, instead of using multiple saturation images as output of the deep-learning model, a single Time of Flight (TOF) image are used to represent CO2 plume propagation; second, we use variational autoencoder-decoder (VAE) for high dimensional image data compression considering uncertainties in the predicted images. Time-of-Flight (TOF) is the travel time of a neural particle from the injection point, which is calculated along streamline based on the velocity field by considering reservoir heterogeneity and driving forces. The latent variables of VAE are estimated through a deep neural network model with inputs of distributed pressure and temperature measurements. Subsequently, the decoder part of the VAE expands the estimated latent variables to the original dimension of the TOF images. We demonstrate the efficacy of the proposed method by comparison with the more traditional pareto-based multi-objective Genetic Algorithm (MOGA) for the assimilation of the field measurements and forecasting of the CO2 plume migration.The power and utility of our proposed workflow are demonstrated by application to the Illinois Basin-Decatur Project (IBDP). The IBDP is a large scale 3-year CO2 storage test and is part of the Midwest Regional Carbon Sequestration Partnership (MRCSP). The field measurements include bottom-hole pressure and distributed temperature sensing (DTS) data at the injection well, and the distributed pressure measurements at several locations along the monitoring well. The dynamic model is a compositional model with 1.7 million cells. First, we identify the influential parameters using sensitivity analysis based on the pressure and temperature responses. Next, the sensitive parameters are sampled from a pre-specified range and a comprehensive training dataset is generated, including the pressure and temperature measurements and TOF images based on flow simulation results from a variety of geological model realizations. The trained neural network model can identify geological model realizations that capture the spatial and temporal migration of pressure and CO2 saturation based on Euclidean distance within the VAE latent space. For comparison with traditional history matching workflow, the selected sensitive parameters are also tuned using MOGA with the pressure and temperature measurements. A reasonable agreement is obtained for the predicted CO2 plume migration between two workflows. The proposed deep learning workflow shows significant speed-up compared to the traditional multi-objective optimization-based history matching workflow (5 h for training and seconds for prediction; compared to several days/weeks for the MOGA).The novelty of this work is the application of an efficient deep learning-based workflow to a large-scale CO2 storage test site for assimilating field measurements and predicting CO2 plume propagation in a near real-time and comparison with traditional history matching approach.

Optimal performance and preliminary parameter matching for hydrogen fuel cell powertrain system of electric aircraft

Fuel cells are true net-zero carbon emission power sources for aircraft, which is highly sensitive to weight. In the initial phase of adapting hydrogen fuel cell systems for aircraft powertrains, preliminary design parameter matching remains premature. An explicit method for the performance optimization of aircraft hydrogen fuel cell powertrain systems and a process of preliminary parameter matching are proposed to address this problem. Performance and weight models of the fuel cell stack and its auxiliaries, the cathode air compressor subsystem, and the cooling subsystem are designed, and system performance at various altitudes and power output levels is calculated. The aircraft flight mission performance is synthesized and considered in the optimization process. The optimization result of system performance and the corresponding design parameters are then graphically illustrated as tern plots. Unlike the traditional iterative preliminary system parameter matching and optimization method, which explores the design space non-directionally and converges to a single local optimal point, the proposed explicit method sweeps the design space globally and obtains a group of design points with acceptable optimality. The system design process is boosted by a compact iterative loop. In the optimization practice, the cruise powertrain specific energy is improved by 6.5%. The relationship between specific system design parameters and system performance is displayed globally by the resulting tern plots. Multiple design guidelines are observed and proposed, and design scenarios are directly obtained from the graphs for further engineering processes.

A cascaded GRU-based stereoscopic matching network for precise plank measurement

Wooden plank images in industrial measurements often contain numerous textureless areas. Furthermore, due to the thin plate structure, the three-dimensional (3D) disparity of these planks is predominantly confined to a narrow range. Consequently, achieving accurate 3D matching of wooden plank images has consistently presented a challenging task within the industry. In recent years, deep learning has progressively supplanted traditional stereo matching methods due to its inherent advantages, including rapid inference and end-to-end processing. Nonetheless, the acquisition of datasets for stereo matching networks poses an additional challenge, primarily attributable to the difficulty in obtaining accurate disparity data. Thus, this paper presents a novel stereo matching method incorporating three key innovations. Firstly, an enhanced gated recurrent unit network is introduced, accompanied by a redesigned structure to achieve higher matching accuracy. Secondly, an efficient preprocessing module is proposed, aimed at improving the algorithm’s efficiency. Lastly, in response to the challenges posed by datasets acquisition, we innovatively employed image simulation software to obtain a high-quality simulated dataset of wooden planks. To assess the feasibility of our approach, we conducted both simulated and real experiments. The experiments results clearly exhibit the superiority of our method when compared to existing approaches in terms of both stability and accuracy. In the simulation experiment, our method attained a bad1.0 score of 2.1% (compared to the baseline method’s 9.76%); In the real experiment, our method achieved an average error of 0.104 mm (compared to the baseline method’s 0.268 mm). It is worth noting that our study aims to address the challenge of acquiring datasets for deep learning and bridging the gap between simulated and real data, resulting in increased applicability of deep learning in more industrial measurement domains.

Making the most of AI and machine learning in organizations and strategy research: Supervised machine learning, causal inference, and matching models

AbstractResearch SummaryWe spotlight the use of machine learning in two‐stage matching models to deal with sample selection bias. Recent advances in machine learning have unlocked new empirical possibilities for inductive theorizing. In contrast, the opportunities to use machine learning in regression studies involving large‐scale data with many covariates and a causal claim are still less well understood. Our core contribution is to guide researchers in the use of machine learning approaches to choosing matching variables for enhanced causal inference in propensity score matching models. We use an analysis of real‐world technology invention data of public–private relationships to demonstrate the method and find that machine learning can provide an alternative approach to ad hoc matching. However, as with any method, it is also important to understand its limitations.Managerial SummaryThis article explores the use of machine learning to enhance decision‐making, particularly in addressing sample selection bias in large‐scale datasets. The rapid development of AI and machine learning offers new, powerful tools especially for digital ecosystems where complex data and causal relationships are complex to analyze. We offer managers and stakeholders insight into the effective integration of machine learning for selecting critical variables in propensity score matching models. Through a detailed examination of real‐world data on technology inventions within public–private relationships, we demonstrate the effectiveness of machine learning as a robust alternative to traditional matching methods.