Adaptive Recognition Research Articles

Semi-supervised graph convolutional networks for the domain adaptive recognition of thyroid nodules in cross-device ultrasound images.

Ultrasound plays a critical role in the early screening and diagnosis of cancers. Although deep neural networks have been widely investigated in the computer-aided diagnosis (CAD) of different medical images, diverse ultrasound devices, and image modalities pose challenges for clinical applications, especially in the recognition of thyroid nodules having various shapes and sizes. More generalized and extensible methods need to be developed for the cross-devices recognition of thyroidnodules. In this work, a semi-supervised graph convolutional deep learning framework is proposed for the domain adaptative recognition of thyroid nodules across several ultrasound devices. A deep classification network, trained on a source domain with a specific device, can be transferred to recognize thyroid nodules on the target domain with other devices, using only few manual annotated ultrasoundimages. This study presents a semi-supervised graph-convolutional-network-based domain adaptation framework, namely Semi-GCNs-DA. Based on the ResNet backbone, it is extended in three aspects for domain adaptation, that is, graph convolutional networks (GCNs) for the connection construction between source and target domains, semi-supervised GCNs for accurate target domain recognition, and pseudo labels for unlabeled target domains. Data were collected from 1498 patients comprising 12 108 images with or without thyroid nodules under three different ultrasound devices. Accuracy, Sensitivity and Specificity were used for the performanceevaluation. The proposed method was validated on six groups of data for a single source domain adaptation task, the mean Accuracy was 0.9719± 0.0023, 0.9928± 0.0022, 0.9353± 0.0105, 0.8727± 0.0021, 0.7596± 0.0045, 0.8482± 0.0092, which achieved better performance in comparison with the state-of-the-art. The proposed method was also validated on three groups of multiple source domain adaptation tasks. In particular, when using X60 and HS50 as the source domain data, and H60 as the target domain, it can achieve the Accuracy of 0.8829± 0.0079, Sensitivity of 0.9757± 0.0001, and Specificity of 0.7894± 0.0164. Ablation experiments also demonstrated the effectiveness of the proposedmodules. The developed Semi-GCNs-DA framework can effectively recognize thyroid nodules on different ultrasound devices. The developed semi-supervised GCNs can be further extended to the domain adaptation problems for other modalities of medicalimages.

Improved domain adaptive rice disease image recognition based on a novel attention mechanism

Due to the competitive and better performance on various benchmarks of computer vision, transfer learning has been widely explored in the task of crop disease image recognition with small samples. However, when there is a large difference in data distribution between target domain and source domain, it is difficult to achieve good results using transfer learning. To solve this problem, this paper took rice disease image recognition with small samples as an example, and proposed a domain adaptive image recognition method based on a novel attention mechanism. First, the weight distribution of rice disease image features in the neural network was optimized, so that the improved attention mechanism can pay more attention to the image features closely related to rice diseases. Second, the attention mechanism was integrated with the domain adaptation network to reduce the difference of feature distribution between the datasets, thereby improving the accuracy of rice disease image recognition. Finally, some comparative experiments were designed to verify the effectiveness of the proposed method. Experimental results showed that the method can effectively solve the problem of low image recognition accuracy caused by large difference in data distribution between target domain and source domain. And when the size of the source domain dataset decreased gradually, the method can still maintain a high and stable accuracy. The image recognition accuracies of 95.25%, 91.50% and 91.25% were achieved on three commonly used domain adaptation networks respectively. The achievement of this paper can offer a new method for crop disease image recognition with small samples.

마스크 착용에 적응적인 얼굴인식 방법

얼굴인식(face recognition)은 스마트 기기의 사용자 인증, 건물 출입 통제, 지능형 감시 시스템 등에 널리 활용되는 생체인식 기술이다. 코로나19 팬데믹 이후로 마스크 착용이 일상화되면서 마스크 착용 얼굴을 식별하는 마스크 얼굴인식 연구에 대한 중요도가 높아지고 있다. 마스크 얼굴인식에서는 테스트 얼굴 이미지는 마스크 착용이 많은 반면 학습용 얼굴 이미지는 마스크 미착용이 많아서 마스크 착용 여부로 인한 불일치가 발생하고 이는 인식 정확도 저하로 이어지기 쉽다. 본 논문에서 제안하는 마스크 얼굴인식 방법은 앞서 언급한 문제를 극복하기 위해 마스크 미착용 학습용 얼굴 이미지들에 마스크를 합성하여 마스크 착용/미착용 쌍으로 증강된 학습셋을 구성한다. 그리고 마스크 착용 여부를 알 수 없는 테스트 얼굴 이미지를 마스크 착용/미착용 쌍의 학습용 얼굴 이미지들의 최적 조합으로 복원(reconstruction)한 뒤(인물 별로 수행) 복원 오류가 최소인 인물 클래스를 찾아 어떤 인물에 해당하는지 최종 판정한다. 제안하는 방법은 테스트 얼굴 이미지의 마스크 착용 여부와 관계 없이 사용할 수 있다는 장점이 있으며 마스크 착용 시 약 82%, 미착용 시 약 75%의 인식 정확도를 달성하였다.

Structural Adaptation of the Single-Stranded DNA-Binding Protein C-Terminal to DNA Metabolizing Partners Guides Inhibitor Design.

Single-stranded DNA-binding protein (SSB) is a bacterial interaction hub and an appealing target for antimicrobial therapy. Understanding the structural adaptation of the disordered SSB C-terminus (SSB-Ct) to DNA metabolizing enzymes (e.g., ExoI and RecO) is essential for designing high-affinity SSB mimetic inhibitors. Molecular dynamics simulations revealed the transient interactions of SSB-Ct with two hot spots on ExoI and RecO. The residual flexibility of the peptide-protein complexes allows adaptive molecular recognition. Scanning with non-canonical amino acids revealed that modifications at both termini of SSB-Ct could increase the affinity, supporting the two-hot-spot binding model. Combining unnatural amino acid substitutions on both segments of the peptide resulted in enthalpy-enhanced affinity, accompanied by enthalpy-entropy compensation, as determined by isothermal calorimetry. NMR data and molecular modeling confirmed the reduced flexibility of the improved affinity complexes. Our results highlight that the SSB-Ct mimetics bind to the DNA metabolizing targets through the hot spots, interacting with both of segments of the ligands.

Reversible Data Hiding in Encrypted Images Based on an Adaptive Recognition Strategy for Blocks

As the rapid development of third-party storage and homomorphic encryption have profoundly stimulated the desire for secure communication, reversible data hiding in encrypted images has received widespread attention, since it allows lossless data conveying and perfect image recovery. In order to obtain secure reversible data hiding with high embedding capacity, a novel block embedding method is proposed, based on an adaptive recognition strategy for combined blocks in the binary image, with which the adjacent identical blocks can be integrated into a combination to reserve more spare bits for data accommodation. Furthermore, a fully reversible data hiding method for grayscale images in the encryption domain is designed. The secret data is hidden into lower bit-planes of the image while the original bits of those embedded lower pixels are recorded into the vacated space of higher bit-planes. The original image can be reconstructed flawlessly as well as the secret data being extracted without errors. To reinforce security, the original image and the secret data are encrypted and scrambled based on sequences generated with the high-dimension chaotic system. Due to its high sensitivity of initial values, the performance such as security and robustness is guaranteed. By comparing the PSNR value of the marked decrypted image and evaluating the quality of the extracted secret image, experimental results demonstrate that the proposed method can obtain higher embedding capacity, achieving 0.2700–0.3924 bpp increment over the state-of-the-art methods, and recover the marked decrypted image with high visual symmetry/quality, and efficiently resist against potential attacks, such as the histogram analysis, differential, brute-force, JPEG attacks, and so on.

A Real-Time Shrimp with and without Shells Recognition Method for Automatic Peeling Machines Based on Tactile Perception

Accurate and automatic real-time recognition of shrimp with and without shells is the key to improve the efficiency of automatic peeling machines and reduce the labor cost. Existing methods cannot obtain excellent accuracy in the absence of target samples because there are too many species of shrimp to obtain a complete dataset. In this paper, we propose a tactile recognition method with universal applicability. First, we obtained tactile data, e.g., the texture and hardness of the surface of the shrimp, through a novel layout using the same type of sensors, and constructed fusion features based on the energy and nonstationary volatility (ENSV). Second, the ENSV features were input to an adaptive recognition boundary model (ARBM) for training to obtain the recognition boundary of shrimp with and without shells. Finally, the effectiveness of the proposed model was verified by comparison with other tactile models. The method was tested with different species of shrimp and the results were 88.2%, 87.0%, and 89.4%, respectively. The recognition accuracy of the overall, shrimp with shells and shrimp without shells verified the generalizability of the proposed method. This method can help to improve the efficiency of automatic peeling machines and reduce the labor cost.

Location Adaptive Motion Recognition Based on Wi-Fi Feature Enhancement

Action recognition is essential in security monitoring, home care, and behavior analysis. Traditional solutions usually leverage particular devices, such as smart watches, infrared/visible cameras, etc. These methods may narrow the application areas due to the risk of privacy leakage, high equipment cost, and over/under-exposure. Using wireless signals for motion recognition can effectively avoid the above problems. However, the motion recognition technology based on Wi-Fi signals currently has some defects, such as low resolution caused by narrow signal bandwidth, poor environmental adaptability caused by the multi-path effect, etc., which make it hard for commercial applications. To solve the above problems, we first propose and implement a position adaptive motion recognition method based on Wi-Fi feature enhancement, which is composed of an enhanced Wi-Fi features module and an enhanced convolution Transformer network. Meanwhile, we improve the generalization ability in the signal processing stage to avoid building an extremely complex model and reduce the demand for system hardware. To verify the generalization of the method, we implement real-world experiments using 9300 network cards and the PicoScenes software platform for data acquisition and processing. By contrast with the baseline method using original channel state information(CSI) data, the average accuracy of our algorithm is improved by 14% in different positions and over 16% in different orientations. Meanwhile, our method has best performance with an accuracy of 90.33% compared with the existing models on public datasets WiAR and WiDAR.

Drop to Transfer: Learning Transferable Features for Robot Tactile Material Recognition in Open Scene

Tactile feedback is an important way for robots to perceive the environment. In the open scenes, many uncertain factors lead to the change of the external environment, such as the re-installation of the robot end-effector and the change of exploratory motions. This situation unavoidably raises the domain shift problem, which makes it difficult for the robot to fit the new open environment if it makes decisions only based on the knowledge extracted from the closed environment. Thus, we propose a new deep unsupervised domain adaptive material recognition model, called Drop To Transfer (DTT), which adaptively learns tactile transferable features to achieve accurate material recognition in open scenes. DTT performs the clustering hypothesis on the target domain samples using the adversarial dropout. And in each training iteration, high-gradient features are progressively removed to enhance the expression of other transferable features. Meanwhile, the Maximum Mean Discrepancy (MMD) metric is added in the optimization process to achieve feature alignment between the source and target domain distributions. In addition, we design an adaptive weighted optimization strategy to optimize each loss in DTT effectively. Experimental results show that the proposed method can effectively learn the transferable features and reduce the domain shift problem for the robot tactile perception task in the open scenes.

Chiral adaptive recognition with sequence specificity of aromatic dipeptides in aqueous solution by an achiral cage.

Sequence-specific recognition of peptides and proteins by synthetic compounds or systems remains a huge challenge in biocompatible media. Here, we report the chiral adaptive recognition (CAR) with sequence specificity of aromatic dipeptides in a purely aqueous solution using an achiral tetraphenylethene-based octacationic cage (1) as both a molecular receptor and chiroptical sensor. 1 can selectively bind and dimerize aromatic dipeptides to form 1 : 2 host-guest complexes with high binding affinity (>1010 M-2), especially up to ∼1014 M-2 for TrpTrp. Given the dynamic rotational conformation of TPE units, achiral 1 can exhibit chiral adaptive responses with mirror-symmetrical circular dichroism (CD) and circularly polarized luminescence (CPL) spectra to enantiomeric dipeptides via supramolecular chirality transfer in the host-guest complexes. Furthermore, this CAR with sequence specificity of 1 can be applied for molecular recognition of TrpTrp- or PhePhe-containing tetrapeptides, polypeptides (e.g., amyloid β-peptide1-20 and somatostatin), and proteins (e.g., human insulin) with characteristic CD responses.

Real-Time Wrist Motion Decoding with High Framerate Electrical Impedance Tomography (EIT).

Human wrist motion decoding with a biological-signal-based interface is a key technique in the upper-limb exoskeleton and prosthesis control. One critical issue in this field is achieving high recognition precision and fast time response while against external disturbances of sensor re-wearing. In this study, we proposed a high-framerate Electrical Impedance Tomography (EIT) system combined with an adaptive recognition algorithm for real-time wrist kinematics decoding. The high-framerate EIT system was developed by a parallel stimulation-measurement sequence, and the sampling rate was as high as 104 Hz. Compared to the most widely used myoelectric techniques, the EIT-based interface can provide extra deep muscular spatial information with similar surface electrodes. It greatly benefited the subsequent recognition algorithms, in which the key EIT regions indicating muscle morphology kept consistent after an arbitrary sensor re-donning. The designed adaptive algorithm achieved equally high performance with an automatic update of the classifier mean values with a fast self-operated calibration process. We validated the approach on 12 subjects with a 2-dimensional Fitts' law test. The wrist gestures and joint angles were mapped to the direction and speed of the cursor movement, respectively. The average throughputs (TPs) of Fitts' law tests were 1.0269 ± 0.0971 bits/s and 1.0095 ± 0.0931 bits/s without and with sensor re-donning, respectively, which were comparable to the TPs of sEMG-based studies. The results showed the promise of the EIT-based interface on real-time human motion intent recognition. Future endeavors are worth being paid in this direction for more complicated robotic tasks.

Adaptive image recognition circuit based on CMOS-based memristor

Adaptive image recognition circuit based on CMOS-based memristor

Non-invasive measurements of thermal discomfort for thermal preference prediction based on occupants' adaptive behavior recognition

With the spread of the Internet of Things in the construction industry, non-invasive thermal adaptive behavior recognition provides a new method for the real-time assessment of the indoor thermal environment. Pioneering research on non-invasive measurements of thermal discomfort for thermal preference prediction based on occupants' adaptive behavior recognition was undertaken. Questionnaire surveys on the thermal adaptive behavior of office building users in five climatic regions of China were undertaken. From 654 valid questionnaires, 14 thermal adaptive behaviors related to thermal discomfort were clustered. Then experiments were conducted to collect thermal discomfort adaptive behavioral video data from 8 views (front, rear, left, right, left-front, right-front, left-rear, and right-rear) at two heights (1.5 m and 2 m). A large-scale dataset for thermal adaptive behavior recognition with 27,504 video samples was developed. Two adaptive behavior recognition models were constructed based on the neural network model Two-Stream Inflated 3D ConvNet (I3D) and SlowFast. The results show that there is no significant difference in the types of indoor thermal adaptive behaviors among office users in different climate zones of China. The established I3D and SlowFast recognition models adopt an end-to-end approach for recognition, which overcomes dependence on human key point recognition and enables the models to utilize more environmental information. The average prediction accuracy of both models reaches 95% or above, even when complex background and human key points are partially obscured. Additionally, the recognition time of the two models is at the millisecond level, which can support real-time thermal adaptive behavior recognition.

Adaptive insertion of a hydrophobic anchor into a poly(ethylene glycol) host for programmable surface functionalization

Covalent and non-covalent molecular binding are two strategies to tailor surface properties and functions. However, the lack of responsiveness and requirement for specific binding groups makes spatiotemporal control challenging. Here, we report the adaptive insertion of a hydrophobic anchor into a poly(ethylene glycol) (PEG) host as a non-covalent binding strategy for surface functionalization. By using polycyclic aromatic hydrocarbons as the hydrophobic anchor, hydrophilic charged and non-charged functional modules were spontaneously loaded onto PEG corona in 2 min without the assistance of any catalysts and binding groups. The thermodynamically favourable insertion of the hydrophobic anchor can be reversed by pulling the functional module, enabling programmable surface functionalization. We anticipate that the adaptive molecular recognition between the hydrophobic anchor and the PEG host will challenge the hydrophilic understanding of PEG and enhance the progress in nanomedicine, advanced materials and nanotechnology.

Unsupervised domain adaptive tumor region recognition for Ki67 automated assisted quantification.

Ki67 is a protein associated with tumor proliferation and metastasis in breast cancer and acts as an essential prognostic factor. Clinical work requires recognizing tumor regions on Ki67-stained whole-slide images (WSIs) before quantitation. Deep learning has the potential to provide assistance but largely relies on massive annotations and consumes a huge amount of time and energy. Hence, a novel tumor region recognition approach is proposed for more precise Ki67 quantification. An unsupervised domain adaptive method is proposed, which combines adversarial and self-training. The model trained on labeled hematoxylin and eosin (H&E) data and unlabeled Ki67 data can recognize tumor regions in Ki67 WSIs. Based on the UDA method, a Ki67 automated assisted quantification system is developed, which contains foreground segmentation, tumor region recognition, cell counting, and WSI-level score calculation. The proposed UDA method achieves high performance in tumor region recognition and Ki67 quantification. The AUC reached 0.9915, 0.9352, and 0.9689 on the validation set and internal and external test sets, respectively, substantially exceeding baseline (0.9334, 0.9167, 0.9408) and rivaling the fully supervised method (0.9950, 0.9284, 0.9652). The evaluation of automated quantification on 148 WSIs illustrated statistical agreement with pathological reports. The model trained by the proposed method is capable of accurately recognizing Ki67 tumor regions. The proposed UDA method can be readily extended to other types of immunohistochemical staining images. The results of automated assisted quantification are accurate and interpretable to provide assistance to both junior and senior pathologists in their interpretation.

Performance optimization of the active noise control algorithm based on logarithmic function and its application in secondary channel online identification ANC system

In this paper, several methods about the active noise control (ANC) system are studied. The common adaptive recognition system based on the variable step filter x least mean square (FxLMS) algorithm are elaborated. By introducing the momentum term into the variable step algorithm, the value range of the variable step is increased, so as to speed up the convergence of the algorithm and improve the effect of noise reduction. The weight of the control filter based on the improved logarithmic function variable step size least mean square (VSSLMS) algorithm is updated in the improved second channel online recognition algorithm. The initial smooth noise signal, the noise signal of increasing amplitude, and the abrupt noise signal are utilized to verify the convergence speed and the noise reduction effect of the modified ANC method. Numerical results show that the improved algorithm structure could improve the convergence speed and reduce the noises of the whole system at three different primary noises. The initial smooth noise signal, the noise signal of increasing amplitude, and the abrupt noise signal are utilized to verify the convergence speed and the noise reduction effect of the modified ANC method. Numerical results under three different primary noises show that the improved algorithm structure could improve the convergence speed and reduce the noise of the whole system.

Adaptive ship-radiated noise recognition with learnable fine-grained wavelet transform

Analyzing the ocean acoustic environment is a tricky task. Background noise and variable channel transmission environment make it complicated to implement accurate ship-radiated noise recognition. Existing recognition systems are weak in addressing the variable underwater environment, thus leading to disappointing performance in practical application. In order to keep the recognition system robust in various underwater environments, this work proposes an adaptive generalized recognition system — AGNet (Adaptive Generalized Network). By converting fixed wavelet parameters into fine-grained learnable parameters, AGNet learns the characteristics of underwater sound at different frequencies. Its flexible and fine-grained design is conducive to capturing more background acoustic information (e.g., background noise, underwater transmission channel). To utilize the implicit information in wavelet spectrograms, AGNet adopts the convolutional neural network with parallel convolution attention modules as the classifier. Experiments reveal that our AGNet outperforms all baseline methods on several underwater acoustic datasets, and AGNet could benefit more from transfer learning. Moreover, AGNet shows robust performance against various interference factors.

A New Design of Iris Recognition Using Hough Transform with K-Means Clustering and Enhanced Faster R-CNN

Iris recognition method is the most significant biometric modality for human identification. Currently, multiple deep structured architectures have been employed in biometric recognition with various features like robust generalization ability, high accuracy, and automatic learning. It is also considered as a more emerging area in biometrics along with machine learning approaches. Deep CNN is a conventional approach of image processing broadly utilized in several fields, however, it is simply influenced by slight disturbances and also includes bad anti-noise capacity in image classification. The major scope of this work is to implement adaptive deep learning-aided iris recognition. The key contribution of this paper relies on enhancing iris segmentation and iris recognition. Initially, the iris images are pre-processed, which are subjected to segmentation. The segmentation approach focuses on a Hough transform with K-Means Clustering (HT-KMC). Once the segmentation of the iris is done, the “Enhanced Faster Region-Convolutional Neural Network (E-Faster-RCNN)” is developed for the recognition purpose. The number of suitably hidden neurons of the conventional Faster-RCNN is optimized by the Adaptive-Sail Fish Optimization (ASFO) algorithm to obtain the highest efficiency. The simulation findings of the recommended approach proved that the method is effective in iris recognition by experimenting on a benchmark dataset.

Design of an Intervention and Education System for Children with Emotional Disorders Based on Semantic Analysis.

In this paper, a semantic analysis approach to children's emotional disorder intervention and education is thoroughly analyzed and discussed, and a corresponding educational system is designed for application in real life. This paper acquires video data by deploying a common camera acquisition and transforms, annotates, frames, and processes the data with the help of feature engineering methods. In addition, this paper proposes a fine-grained action decomposition strategy to solve the problem of extreme imbalance in the dataset to improve the performance of the model and proposes an iterative sampling data fusion strategy, which aims to integrate and fuse data from multiple sources to make them more effective and further improve the robustness and generalization ability of the model. Since it is difficult for families to improve the emotional management skills of migrant children, and it is also difficult to obtain professional help and support from the community or schools, it is important to take advantage of the professional strengths of social work to provide professional support for migrant children and their families. From the perspective of theoretical research, most of the existing studies focus on individual migrant children and cannot give global guidance from the perspective of the family system. The comparison results show that T-SVR trained using data from all subjects outperforms the inductive method based on individual training of trainees, validating the effectiveness of the proposed adaptive emotion recognition model. Therefore, from the perspective of system integration, it is important to explore social work interventions to improve the emotional management skills of migrant children. The system network structure design is determined according to the actual situation; then from the system requirements, the system is abstracted with the help of UML entity-relationship diagram, and the database table design is completed; so far, the overall system can be divided into independent functional modules, and the boundaries of each module and the participating roles are gradually clarified, and the detailed design within each functional module is illustrated by UML timing diagram and class diagram to clarify the classes used. Finally, the system is tested end-to-end to verify whether the results of the view layer meet the design guidelines, whether the system modules work together properly, and whether the functional development meets the requirements.

An Outlier Cleaning Based Adaptive Recognition Method for Degradation Stage of Bearings.

Accurate identification of the degradation stage is key to the prediction of the remaining useful life (RUL) of bearings. The 3σ method is commonly used to identify the degradation point. However, the recognition accuracy is seriously disturbed by the random outliers in the normal stage. Therefore, this paper proposes an adaptive recognition method for the degradation stage based on outlier cleaning. Firstly, an improved multi-scale kernel regression outlier detection method is adopted to roughly search the abnormal signal segments. Then, a method for the accurate locating of the start and end points of abnormal impulses is established. After that, indexes are constructed for screening abnormal segments and an iterative strategy is proposed to achieve an accurate and efficient removal of abnormal impulses. After outlier cleaning, the 3σ approach is used to set the degradation warning threshold adaptively to realize the degradation stage recognition of the bearings. The PHM 2012 rotating machinery dataset is used to verify the effectiveness of the proposed method. Experimental results show that the proposed method can accurately locate and remove the outliers adaptively. After the cleaning of the outliers, the identification of the degradation stage is no longer disturbed by the selection of the reference signal of the normal stage and the robustness and the accuracy of the degradation stage identification have been improved significantly.

A Strategy for Variable-Scale InSAR Deformation Monitoring in a Wide Area: A Case Study in the Turpan–Hami Basin, China

In recent years, increasing available synthetic aperture radar (SAR) satellite data and gradually developing interferometric SAR (InSAR) technology have provided the possibility for wide-scale ground-deformation monitoring using InSAR. Traditionally, the InSAR data are processed by the existing time-series InSAR (TS–InSAR) technology, which has inefficient calculation and redundant results. In this study, we propose a wide-area InSAR variable-scale deformation detection strategy (hereafter referred to as the WAVS–InSAR strategy). The strategy combines stacking technology for fast ground-deformation rate calculation and advanced TS–InSAR technology for obtaining fine deformation time series. It adopts an adaptive recognition algorithm to identify the spatial distribution and area of deformation regions (regions of interest, ROI) in the wide study area and uses a novel wide-area deformation product organization structure to generate variable-scale deformation products. The Turpan–Hami basin in western China is selected as the wide study area (277,000 km2) to verify the proposed WAVS–InSAR strategy. The results are as follows: (1) There are 32 deformation regions with an area of ≥1 km2 and a deformation magnitude of greater than ±2 cm/year in the Turpan–Hami basin. The deformation area accounts for 2.4‰ of the total monitoring area. (2) A large area of ground subsidence has occurred in the farmland areas of the ROI, which is caused by groundwater overexploitation. The popularization and application of facility agriculture in the ROI have increased the demand for irrigation water. Due to the influence of the tectonic fault, the water supply of the ROI is mainly dependent on groundwater. Huge water demand has led to a continuous net deficit in aquifers, leading to land subsidence. The WAVS–InSAR strategy will be helpful for InSAR deformation monitoring at a national/regional scale and promoting the engineering application of InSAR technology.