Candidate Areas Research Articles

An commercial area extraction approach using time series nighttime light remote sensing data——Take Wuhan city as a case

The degree of nighttime illumination in a region is closely related to its economic development, therefore, urban commercial areas can be located based on the intensity of the nighttime light. Aiming at the problem of discontinuous and incomparable brightness of nighttime remote sensing images in different periods, a method of urban commercial area extraction based on temporal nighttime remote sensing data is proposed. Firstly, multi- nighttime remote sensing image correction was carried out based on the invariant target region method, and stable time series NPP/VIIRS data set was generated. Secondly, the impervious water surface was extracted by three index synthesis method to obtain the candidate urban commercial area using Landsat-8 image. Finally, multi-scale segmentation and unsupervised classification are carried out on the temporal nighttime remote sensing data set of the candidate area to realize the extraction of urban commercial area. Taking Wuhan city as an example, the method in this paper is used to extract and verify the urban commercial area. The effectiveness and accuracy of the method are verified according to the commercial planning map of Wuhan city and previous studies.

Research on Obstacle Detection and Avoidance of Autonomous Underwater Vehicle Based on Forward-Looking Sonar.

Due to the complexity of the ocean environment, an autonomous underwater vehicle (AUV) is disturbed by obstacles when performing tasks. Therefore, the research on underwater obstacle detection and avoidance is particularly important. Based on the images collected by a forward-looking sonar on an AUV, this article proposes an obstacle detection and avoidance algorithm. First, a deep learning-based obstacle candidate area detection algorithm is developed. This algorithm uses the You Only Look Once (YOLO) v3 network to determine obstacle candidate areas in a sonar image. Then, in the determined obstacle candidate areas, the obstacle detection algorithm based on the improved threshold segmentation algorithm is used to detect obstacles accurately. Finally, using the obstacle detection results obtained from the sonar images, an obstacle avoidance algorithm based on deep reinforcement learning (DRL) is developed to plan a reasonable obstacle avoidance path of an AUV. Experimental results show that the proposed algorithms improve obstacle detection accuracy and processing speed of sonar images. At the same time, the proposed algorithms ensure AUV navigation safety in a complex obstacle environment.

REAL-TIME SMOKE DETECTION IN VIDEO

The paper considers algorithmic and software for early fire detection based on smoke detection from video sequences
 generated by a static video camera. To detect areas with smoke, an algorithm has been developed that allows
 you to select such areas on video frames that are characterized by a number of features: the presence of a stable
 directional movement, compliance with the color characteristics of smoke, and a decrease in the energy value of highfrequency
 components relative to the background model. The feature of the algorithm is a step-by-step spatiotemporal
 analysis of candidate areas, which provides satisfactory computational costs and real-time operation
 on modern computing tools for high-resolution video frames. The algorithm is implemented using the functions
 of the OpenCV computer vision library and multi-threaded processing. The features and main functionality of the software
 implemented as a stationary application are given. The results of experimental studies on the evaluation
 of the efficiency of the algorithm and its speed are presented.

MSER-Vertical Sobel for Vehicle Logo Detection

Detecting a vehicle logo is the first step before realizing the identity of the logo. However, the detection of logos can pose difficulties due to various factors, including logo variations, differing scales and orientations, background interference, varying lighting conditions, and partial obstruction. This paper presents a vehicle logo detection method using hand-crafted features. We used a combination of Maximally Stable Extremal Region (MSER) and Vertical Sobel. We combine vertical Sobel with MSER to overcome MSER's limitation in recognizing objects of different sizes. These two features are merged using a closing morphology operation to form blobs selected as logo candidate areas. Moreover, a Support Vector Machine (SVM) is implemented to choose a logo area by analyzing each candidate's Histogram of Oriented Gradient (HOG). The proposed method was compared with other methods by implementing them on the same dataset. The significant advantage of using MSER-Vertical Sobel is its fast computation time. It is faster than other approaches that use non-handcrafted features. The test results show that the MSER-Vertical Sobel can achieve high accuracy and the fastest computation time.

Infrared small target detection based on multi-perception of target features

The rapid and efficient detection of infrared dim and small targets in complex backgrounds is a critical area of research in infrared image processing. When confronted with limited target features, it becomes vital to mine and construct additional features in order to accurately differentiate the target from the surrounding background and noise. This paper presents a novel small infrared target detection method based on multi-perception of target features which involves four main stages. First, two new target features are constructed to reduce the similarity between the target and noise and smooth the background. Second, two designed filters are applied to enhance the significance of the target and achieve preliminary detection. Third, structure tensor analysis is used to remove the significant background edge regions and perceive targets with strong corner characteristics. Finally, pseudo targets are eliminated by utilizing target candidate areas to achieve the final detection. Experimental results demonstrate that this method is faster and more robust compared to existing detection algorithms, with superior adaptability and detection performance.

Large dam candidate region identification from multi-source remote sensing images via a random forest and spatial analysis approach

ABSTRACTThe extraction of large dam candidate regions is critical for broad-scale efforts to rapidly detect large-area dams. The framework proposed in this paper attempts to combine random forest classification models and spatial analysis methods with large dam candidate area extraction methods for large-scale areas. First, we studied the combination of optical, microwave, texture, and topographic features of the dam and constructed a multisource remote sensing and topographic feature vector of the dam. Secondly, we constructed random forest classifiers in different study areas and evaluate their performance. Then we explored the geographic characteristics of the dams and their relationships with other features. Finally, we introduced the spatial analysis method to constrain the large dam candidate area. The proposed framework was tested in a total area of 968,533 km2 in five countries and achieved promising results, which constrained the candidate area to less than 1.06% of the total area. We calculated the completeness rate of large dams using the multi-source dam datasets. The framework achieved a completeness rate of more than 97.62%. Our results show that the entire framework is reliable for automated and fast large dam candidate area acquisition based on data from open remote sensing products.

Power Line Extraction and Tree Risk Detection Based on Airborne LiDAR.

Transmission lines are the basis of human production and activities. In order to ensure their safe operation, it is essential to regularly conduct transmission line inspections and identify tree risk in a timely manner. In this paper, a power line extraction and tree risk detection method is proposed. Firstly, the height difference and local dimension feature probability model are used to extract power line points, and then the Cloth Simulation Filter algorithm and neighborhood sharing method are creatively introduced to distinguish conductors and ground wires. Secondly, conductor reconstruction is realized by the approach of the linear-catenary model, and numerous non-risk points are excluded by constructing the tree risk point candidate area centered on the conductor's reconstruction curve. Finally, the grading strategy for the safety distance calculation is used to detect the tree risk points. The experimental results show that the precision, recall, and F-score of the conductors (ground wires) classification exceed 98.05% (97.98%), 99.00% (99.14%), and 98.58% (98.56%), respectively, which presents a high classification accuracy. The Root-Mean-Square Error, Maximum Error, and Minimum Error of the conductor's reconstruction are better than 3.67 cm, 7.13 cm, and 2.64 cm, respectively, and the Mean Absolute Error of the safety distance calculation is better than 6.47 cm, proving the effectiveness and rationality of the proposed tree risk points detection method.

Public acceptance of nuclear waste disposal sites: a decision-making process utilising the ‘veil of ignorance’ concept

This study demonstrates that a decision-making process utilising ‘the veil of ignorance’ concept, defined in process terms as beginning from a blank slate encompassing the entire country as potential sites and shortlisting candidate sites based on scientific (geological) safety, promotes public acceptance of siting a repository for the geological disposal of high-level radioactive waste and fosters procedural fairness. A hypothetical scenario experiment was conducted in Japan, manipulating the site selection process by setting two conditions—one being the application/proposal condition that the Japanese government currently employs, such as an application by municipalities or a proposal by the government, and the other being the veil of ignorance condition, in which multiple candidate areas are selected from a blank slate for the entire land area based purely on geological factors. Three stages of acceptance were presumed—at the level of general management policy, the site selection process itself with a specified decision policy, and the siting of a repository in their area of residence. Two hypotheses were tested: (a) the veil of ignorance condition will be evaluated as a more acceptable and fairer procedure and will engender increased national consensus than the application/proposal condition at the site selection and repository siting stages, and (b) procedural fairness and national consensus will impact acceptance at each stage; these variables at each stage help shape the same variables in the next stage. The results supported these hypotheses. This study discusses the importance of the site selection process, beginning where any de facto site can be a candidate and shortlisting the candidate sites based on scientific criteria.

Mesophotic benthic communities associated with a submerged palaeoshoreline in Western Australia.

Key ecological features (KEFs) are elements of Australia's Commonwealth marine environment considered to be important for biodiversity or ecosystem function, yet many KEFs are poorly researched, which can impede effective decision-making about future development and conservation. This study investigates a KEF positioned over the Last Glacial Maximum (LGM) shoreline on the northwest shelf of Australia (known as the 'Ancient Coastline at ~125m depth contour'; AC125). Seafloor bathymetry, sedimentology and benthic habitats were characterised within five study areas using multibeam sonar, sediment samples and towed video imagery. Direct evidence for the existence of a palaeoshoreline formed during the LGM was not found, however candidate areas to find palaeoshoreline material at or just below the modern seabed were discovered. Approximately 98% of the seabed surveyed was comprised of unconsolidated soft sediment habitat (mud/sand/silt) supporting negligible epibenthic biota. The prevalence of soft sediment suggests that post-glacial sediments have infilled parts of the palaeoshoreline, with cross-shelf, probably tidal currents in the northern section of the study area responsible for some of the sediment mobilisation and southern study areas more influenced by oceanic conditions. Within study areas, total biotic cover ranged from 0.02% to 1.07%. Of the biota encountered, most comprised filter feeder organisms (including gorgonians, sponges, and whip corals) whose distribution was associated with pockets of consolidated hard substrate. Benthic community composition varied with both study area and position in relation to the predicted AC125. In general, consolidated substrate was proportionally higher in water shallower than the AC125 compared to on the AC125 or deeper than the AC125. Spatially continuous maps of predicted benthic habitat classes (pre-determined benthic communities) in each study area were developed to characterise biodiversity. Spatial modelling corroborated depth and large-scale structural complexity of the seafloor as surrogates for predicting likely habitat class. This study provides an important assessment of the AC125 and shows that if a distinct coastline exists in the areas we surveyed, it is now largely buried and as such does not provide a unique hard substrate habitat. However, much work remains to fully locate and map the ancient coastline within the vast region of the AC125 and additional surveys in shallow waters adjacent to the AC125 may identify whether some sections lie outside the currently defined KEF.

Underwater object tracker: UOSTrack for marine organism grasping of underwater vehicles

A visual single-object tracker is an indispensable component of underwater vehicles (UVs) in marine organism grasping tasks. Its accuracy and stability are imperative to guide the UVs to perform grasping behavior. Although single-object trackers show competitive performance in the challenge of underwater image degradation, there are still issues with sample imbalance and exclusion of similar objects that need to be addressed for application in marine organism grasping. This paper proposes Underwater OSTrack (UOSTrack), which consists of underwater image and open-air sequence hybrid training (UOHT), and motion-based post-processing (MBPP). The UOHT training paradigm is designed to train the sample-imbalanced underwater tracker so that the tracker is exposed to a great number of underwater domain training samples and learns the feature expressions. The MBPP paradigm is proposed to exclude similar objects. It uses the estimation box predicted with a Kalman filter and the candidate boxes in the response map to relocate the lost tracked object in the candidate area. UOSTrack achieves an average performance improvement of 4.41% and 7.98% maximum compared to state-of-the-art methods on various benchmarks, respectively. Field experiments have verified the accuracy and stability of our proposed UOSTrack for UVs in marine organism grasping tasks. More details can be found at https://github.com/LiYunfengLYF/UOSTrack.

Deep learning for check dam area extraction with optical images and digital elevation model: A case study in the hilly and gully regions of the Loess Plateau, China

AbstractCheck dams are widely used on the Loess Plateau of China to control soil and water loss, develop agricultural land and improve watershed ecology. Detailed information on the spatial distribution of check dams and the area of dam land is critical for quantitatively evaluating hydrological and ecological effects, planning the construction of new dams and repairing damaged dams. Therefore, this research presents a method that integrates deep learning and geospatial analysis to facilitate the extraction of check dam areas in broad areas from high‐resolution Gaofen‐2 (GF‐2) multispectral imageries, including red (R), green (G), blue (B) and near‐infrared (NIR) bands and digital elevation model (DEM). First, we generated three datasets with different band combinations (RGB, RGB + NIR and RGB + DEM) using GF‐2 remote sensing images combined with Advanced Land Observing Satellite—the Phased Array type L‐band Synthetic Aperture Radar DEM data to determine the optimal data combination for dam area extraction. Next, four widely used semantic segmentation networks—Fully Convolutional Network (FCN), U‐Net, PSPNet and DeepLabv3+—were modified to support the arbitrary number of input channels and evaluated for check dam area extraction. Finally, the check dam candidate areas in the Yan River basin were extracted from DEM using geospatial analysis to optimize the dam area extraction results. The results showed that all deep learning (DL) models could extract dam areas quickly and accurately with mean intersection over union and overall accuracy values >85% and >98%, respectively. PSPNet had the best performance for testing datasets with different band combinations. We also found that the DL models in the RGB + DEM images had the best remote sensing image segmentation results, avoiding many miss‐classified pixels. The F1 scores in the RGB + DEM test dataset for FCN U‐Net, PSPNet and DeepLabv3+ reached up to 92%, or 3.3%, 2.3%, 1.9% and 3.1% higher than the corresponding values for RGB images. Potential check dam candidate regions (~3969 km2) were obtained for application analysis, which reduced the original area by 50%. The DL models for the RGB + DEM images were applied in the dam candidate regions, extracting 91.9 km2 of agricultural production dam land and 10.6 km2 of runoff and sediment silted dam land. The extraction results will facilitate quantitative analyses of check dams, improve the management of these structures and promote the efficiency of controlling soil losses.

Identifying epileptogenic abnormalities through spatial clustering of MEG interictal band power.

Successful epilepsy surgery depends on localizing and resecting cerebral abnormalities and networks that generate seizures. Abnormalities, however, may be widely distributed across multiple discontiguous areas. We propose spatially constrained clusters as candidate areas for further investigation and potential resection. We quantified the spatial overlap between the abnormality cluster and subsequent resection, hypothesizing a greater overlap in seizure-free patients. Thirty-four individuals with refractory focal epilepsy underwent pre-surgical resting-state interictal magnetoencephalography (MEG) recording. Fourteen individuals were totally seizure-free (ILAE 1) after surgery and 20 continued to have some seizures post-operatively (ILAE 2+). Band power abnormality maps were derived using controls as a baseline. Patient abnormalities were spatially clustered using the k-means algorithm. The tissue within the cluster containing the most abnormal region was compared with the resection volume using the dice score. The proposed abnormality cluster overlapped with the resection in 71% of ILAE 1 patients. Conversely, an overlap only occurred in 15% of ILAE 2+ patients. This effect discriminated outcome groups well (AUC = 0.82). Our novel approach identifies clusters of spatially similar tissue with high abnormality. This is clinically valuable, providing (a) a data-driven framework to validate current hypotheses of the epileptogenic zone localization or (b) to guide further investigation.

Fine-Scale Species Distribution Modeling of Abies koreana across a Subalpine Zone in South Korea for In Situ Species Conservation

Severe declines in the population of Abies koreana, a conifer native to the subalpine regions of South Korea, are likely a consequence of climate change. However, local-scale modeling of the species’ spatial distribution has seen limited application to in situ conservation policies. Therefore, we intended for this study to examine the applicability of fine-scale species distribution modeling of A. koreana in the Mt. Jiri National Park area in S. Korea in order to explore candidate areas for its in situ conservation. We simulated the potential habitat of the species in the area with four separate models using different dominance patterns, then created an index based on habitability probabilities and residual durations to determine priority conservation areas. Under the highest sensitivity of potential habitats to temperature (spatially downscaled based on geomorphological characteristics), rapid habitat reduction occurred under climate warming in all experiments. At the regional scale, hydrological characteristics such as precipitation and slope characterized different secondary habitat distributional patterns among the experiments. Final conservation priority sites were identified based on specified criteria for the designed index. Our results suggest that a fine-scale modeling system with adequate spatial resolution of environmental inputs is advantageous in representing local habitat characteristics of A. koreana and can be applied to in situ conservation strategies.

Research of object detection algorithms based on CNN

This study first presents a taxonomy of region-based and regression-based target identification methods, and then it evaluates each individual technique in turn. The former can extract picture characteristics from the CNN backbone and apply sliding window procedures to candidate areas to lessen computational complexity. The latter immediately takes data from the convolution layer for feature extraction, target classification, and location regression rather than going via the intermediate layer to retrieve candidate areas. Although the accuracy is only marginally poor, this approach can significantly cut the computation time. Next, we categorize and evaluate the method for detecting Anchor-free targets. First, an appropriate illustration of the target identification technique based on intense prediction is shown. It is also clear that this method, when compared to others, has a high rate of calculation and a low rate of accuracy. The newly suggested corner pooling, Cascade corner pooling, and center pooling have significantly increased the calculation accuracy of the target identification method based on key point estimate. Finally, Transformer, which is very inclusive and powerful, can be effectively used in the field of natural language processing and computer vision.

Survey‐based evaluation of national practices for postgraduate year one pharmacy resident emergency response training

AbstractIntroductionMedical emergency response teams optimally involve interdisciplinary expertise, including pharmacists. A pharmacist may make therapy recommendations, prepare medications, and act as a drug information resource, which are important skills for a postgraduate year one pharmacy resident (PGY1) to develop. Limited data exist evaluating current national practices for PGY1 emergency response.ObjectivesThe purpose of this study was to describe national practices for PGY1 emergency response. These data may help PGY1 programs identify potential best practices for self‐improvement.MethodsAn anonymous survey was developed to describe PGY1 emergency response within the United States and electronically distributed to PGY1 residency program directors. The questionnaire included general demographic information along with questions involving pharmacist and pharmacy resident emergency response and training within their institution. A univariate analysis was performed to evaluate potential predictors of survey respondent confidence in PGY1 independent emergency response by the end of the residency year.ResultsA total of 178 completed responses were obtained (response rate: 33.0%). The majority of respondents had both pharmacist and PGY1 response at their institution, with response to cardiopulmonary arrest being the most common. Of the 36 programs that did not have pharmacy resident involvement, the most common reason was due to lack of pharmacy response at their institution (61.1%). The majority of programs (68.3%) were confident in the ability for a PGY1 to independently respond by the end of their residency year. Only the number of emergency response by the end of the residency year was a predictor in their confidence of PGY1 independent response (p = 0.002).ConclusionThe majority of PGY1 programs had both pharmacist and pharmacy resident response to medical emergencies. The results of this survey may help PGY1 programs identify potential areas of improvement and help prospective residency candidates identify program characteristics that may best prepare them for emergency response.

Watermarking approach based on Hermite transform and a sliding window algorithm.

Nowadays, the distribution of huge amounts of medical images through open networks in telemedicine applications has become increasingly faster and easier. Therefore, a number of considerations are introduced related to the risks of the illegal use of these images, as total diagnosis depends on them. Indeed, the patient's data management, storage, and transmission require a technique for boosting security, integrity and privacy measures in telehealthcare services. In fact, in our previous works, we used polynomial decompositions such as Chebychev orthogonal polynomial transform in medical image watermarking. We then customise our tools for finding the best candidate area for embedding the watermark, always seeking to provide the best solution to this issue. In this research, a variation to medical image watermarking approach based on Hermite transform (HT) is suggested for reliable management of medical data. In this approach, the HT is employed as a preprocessing step so as to extract the texture component of the host medical image. Afterward, a sliding window technique is done to select the most suitable regions for watermark embedding. Lastly, the Arnold transform is utilized for encrypting the watermark to strengthen the security of our scheme. Experiments were conducted on various modalities of medical images. Results indicate that the proposed scheme is robust when subjected to various attacks while preserving a high level of security and invisibility factors. Also, our method preserves the quality of medical images with a good embedding capacity. The obtained results support the use of Hermite Polynomials for the implementation of watermarking in the medical imaging context.

A planning framework to mitigate localized urban stormwater inlet flooding using distributed Green Stormwater Infrastructure at an urban scale: Case study of Dallas, Texas

Mitigation of localized pluvial flooding is one of the major resiliency goals in urban environments, and Green Stormwater Infrastructure (GSI) has the potential to deliver such an outcome. However, there is a lack of systematic approaches to prioritize investment in different candidate areas. This study provides a framework to identify vulnerable stormwater drainage inlets and their contributing areas, prioritize them, identify dominant factors in their selection, assess the potential of GSI in mitigating their overflows, and compare the impact and its cost to gray infrastructure upgrade alternatives. Using SWMM 5.1.013, decision trees, and a volumetric-based assessment of GSI overflow capture, we applied the framework to the City of Dallas, Texas, for three design storms with three GSI practices— bioretention cells, raingardens, and rainwater harvesting tanks. Results showed that there was a significant increase in the number of overflowing stormwater drainage inlets, referred to as hotspots, and their contributing subwatersheds, referred to as opportunity areas, with more intense storms especially in problematic watersheds. Also, prioritization results provided a series of maps to rank the opportunity areas based on overflow severity, recurrence of the overflows, and GSI availability. Moreover, classification results showed that inlet features, especially the inlet depth, were the dominant factors in the identification of the non-problematic inlets. Finally, the GSI impact assessment showed substantial overflow mitigation even at the “very high” severity levels when GSI is comprehensively deployed across opportunity areas. Despite gray infrastructure upgrades yielding higher reduction levels, their cost per cubic meter was higher than GSI. Therefore, a combination of GSI and gray results in maximum overflow reduction at a lower cost compared to common practices.

Protected area planning to conserve biodiversity in an uncertain future.

Protected areas are a key instrument for conservation. Despite this, they are vulnerable to risks associated with weak governance, land-use intensification, and climate change. We used a novel hierarchical optimization approach to identify priority areas for expanding the global protected area system that explicitly accounted for such risks while maximizing protection of all known terrestrial vertebrate species. To incorporate risk categories, we built on the minimum set problem, where the objective is to reach species distribution protection targets while accounting for 1 constraint, such as land cost or area. We expanded this approach to include multiple objectives accounting for risk in the problem formulation by treating each risk layer as a separate objective in the problem formulation. Reducing exposure to these risks required expanding the area of the global protected area system by 1.6% while still meeting conservation targets. Incorporating risks from weak governance drove the greatest changes in spatial priorities for protection, and incorporating risks from climate change required the largest increase (2.52%) in global protected area. Conserving wide-ranging species required countries with relatively strong governance to protect more land when they bordered nations with comparatively weak governance. Our results underscore the need for cross-jurisdictional coordination and demonstrate how risk can be efficiently incorporated into conservation planning. Planeación de las áreas protegidas para conservar la biodiversidad en un futuro incierto.

Model-Driven Precise Degradation Analysis Method of Highway Marking Using Mobile Laser Scanning Point Clouds

Highway markings (HMs) are representative elements of inventory digitalization in highway scenes. The accurate position, semantics, and maintenance information of HMs provide significant support for the intelligent management of highways. This article presents a robust and efficient approach for extracting, reconstructing, and degrading analyzing HMs in complex highway scenes. Compared with existing road marking extraction methods, not only can extract HMs in presence of wear and occlusion from point clouds, but we also perform a degradation analysis for HMs. First, the HMs candidate area is determined accurately by sophisticated image processing. Second, the prior knowledge of marking design rules and edge-based matching model that leverages the standard geometric template and radiometric appearance of HMs is used for accurately extracting and reconstructing solid lines and nonsolid markings of HMs, respectively. Finally, two degradation indicators are constructed to describe the completeness of the marking contour and consistency within the marking. Comprehensive experiments on two existing highways revealed that the proposed methods achieved an overall performance of 95.4% and 95.4% in the recall and 93.8% and 95.5% in the precision for solid line and nonsolid line markings, respectively, even with imperfect data. Meanwhile, a database can be established to facilitate agencies' efficient maintenance.

Improved Adaboost-Camshift Face Tracking System in Complex Background

With the rapid growth of science and technology, people pay more and more attention to pattern recognition and computer interaction. Therefore, in the last few years, face detection and tracking technology in video sequence has become a hot topic for people to study. Face tracking detection has a wide application prospect in human-computer interaction, intelligent monitoring, video conference and other aspects. In this paper, the problem of face tracking in video sequences is divided into two aspects: face detection and moving object tracking algorithm. In the face detection problem, the face detection based on Adaboost algorithm is described in detail, and the three-frame difference method is added to make the algorithm better and enhance the speed of face detection. In terms of moving object tracking algorithm, Camshift face tracking algorithm based on color histogram is adopted, which is not affected by the shape and size of the target and has good real-time performance. However, under the influence of color interference and occlusion, the algorithm will make tracking errors. Therefore, Kalman filter is introduced. The algorithm can directly delineate the candidate areas of face to be detected, so as to ensure the feasibility of face tracking. The simulation video image face tracking system is verified by Matlab software. The experimental results show that the system can accurately detect and track the faces in the video image sequence, not only in the simple background, but also in the complex background and multiple faces can also be well detected and tracked, and the tracking ensures real-time performance.