Visual Research Articles

Artificial Visual Network with Fully Modeled Retinal Direction-Selective Neural Pathway for Motion Direction Detection in Grayscale Scenes

The complexity and functional evolution of mammalian visual systems have always been a focal point in neuroscience and biological science research. The primary neurons that output motion direction signals have been a focal point of research in visual neuroscience for nearly 130 years. These neurons are widely present in the cortex and retina of mammals. Although the relevant pathways have been discovered and studied for almost 60 years due to experimental accessibility, research still remains at the cellular level. The specific functions and overall operational mechanisms of the component neurons in the motion direction-selective pathways are yet to be clearly elucidated. In this study, we modeled existing relevant neuroscience conclusions based on the symmetry and asymmetry of whole cells in the retina-to-cortex pathway and proposed a quantitative mechanism for motion direction selectivity pathways, called the Artificial Visual System (AVS). By tests based on 1 million instances of 2D, eight-direction grayscale moving objects, including 10 randomly shaped objects of various sizes, we confirm AVS’s high effectiveness on motion direction detecting. Furthermore, by comparing the AVS with two well-known convolutional neural networks, namely LeNet-5 and EfficientNetB0, we verify its efficiency, generalization, and noise resistance. Moreover, the analysis indicates that the AVS exhibits evident biomimetic characteristics and application advantages concerning hardware implementation, biological plausibility, interpretability, parameter count, and learning difficulty.

Specifics of the Development of an On-Board Visualization System for Civil Aircrafts

Specifics of the Development of an On-Board Visualization System for Civil Aircrafts

Two coacting shadow enhancers regulate twin of eyeless expression during early Drosophila development.

The Drosophila PAX6 homolog twin of eyeless (toy) sits at the pinnacle of the genetic pathway controlling eye development, the retinal determination network. Expression of toy in the embryo is first detectable at cellular blastoderm stage 5 in an anterior-dorsal band in the presumptive procephalic neuroectoderm, which gives rise to the primordia of the visual system and brain. Although several maternal and gap transcription factors that generate positional information in the embryo have been implicated in controlling toy, the regulation of toy expression in the early embryo is currently not well characterized. In this study, we adopt an integrated experimental approach utilizing bioinformatics, molecular genetic testing of putative enhancers in transgenic reporter gene assays and quantitative analysis of expression patterns in the early embryo, to identify 2 novel coacting enhancers at the toy gene. In addition, we apply mathematical modeling to dissect the regulatory landscape for toy. We demonstrate that relatively simple thermodynamic-based models, incorporating only 5 TF binding sites, can accurately predict gene expression from the 2 coacting enhancers and that the HUNCHBACK TF plays a critical regulatory role through a dual-modality function as an activator and repressor. Our analysis also reveals that the molecular architecture of the 2 enhancers is very different, indicating that the underlying regulatory logic they employ is distinct.

Study on microscopic production characteristics of crude oil in tight reservoirs displaced by CO2-assisted fracturing fluid

The CO2-assisted fracturing fluid displacement technology is of great significance for enhancing oil recovery (EOR) of tight reservoirs, while the characteristics and mechanism of microscopic production for CO2-assisted fracturing fluid are still unclear. In this study, the displacement experiments of fracturing fluid and CO2-assisted fracturing fluid were launched with two various methods, the nuclear magnetic resonance technology (NMR) and high-temperature and -pressure visualization system respectively. The experimental results show that crude oil in mesopores and macropores is mainly displaced by fracturing fluids, and the residual oil is mainly retained in micropores and small pores. CO2-assisted fracturing fluid improved the oil recovery of tight reservoirs, with more crude oil recovered from small pores, mesopores and macropores, and less residual oil content in the form of flake and droplet. This study provides a prospective EOR technology for the efficient development of tight reservoirs.

Development of a Myogenin minimal promoter-based system for visualizing the degree of myogenic differentiation

Myogenic differentiation plays a fundamental role in myogenesis during development and in muscle regeneration. Sequential expression of myogenic regulatory factors (MRFs) including myogenin in the progenitor cells triggers the expression of effector proteins such as myosin heavy chain (MHC), leading to the terminal muscle differentiation. Although we have a snapshot-like understanding of molecules at each stage of the differentiation, how these molecules are interrelated in the continuum of myogenic differentiation remains poorly understood. In this study, we analyzed the dynamics of the minimal Myogenin promoter activity in live myoblasts. With the development of a new co-expression analysis method, we were able to reveal in detail the relationship between this Myogenin promoter activity and the expression of endogenous myogenin or MHC, as differentiation markers. Consequently, we found that our visualization system of myogenic differentiation is suitable for monitoring the transition from myoblasts to myotubes, in which the Myogenin promoter activity quantitatively represents the degree of myogenic differentiation. Thus, this system allows simultaneous observation of the degree of myoblast differentiation in relation to other molecules, which would contribute to deepening our understanding of myogenic differentiation as a continuous process.

Regulatory Genes in Eyespot Formation and Function of Mytilus coruscus.

Light sensitivity is important for marine benthic invertebrates, and it plays a vital role in the marine bivalves settling. Animal visual systems are enormously diverse; their development appears to be controlled by a set of conserved retinal determination genes (RDGs). Eyespots, as the simplest animal eyes, their appearance indicates the important effect on mussel larvae attachment. Nevertheless, the molecular mechanism of the eyespot's development in Mytilus coruscus larvae is not clear. In this study, we identified 11 genes which play a regulatory role in the visual system (i.e. Pax1/9, Pax2/5/8, Pax6, Pax3/7, Six1/2, Six3/6, Six4/5, Dach, Eya, Brn and Tbx2) from transcriptome data and the whole genome sequence of M. coruscus. The results of chromosome localization showed that 11 genes were distributed on different chromosomes. Subcellular mapping revealed that all the proteins except Brn were located in the nucleus. Phylogeny and gene structure analyses revealed that the Pax members were divided into four subfamilies, the Six members were divided into three subfamilies and structures within the same subfamily were relatively conserved. Quantitative real-time PCR (qPCR) showed that Dach, Pax6, Pax3/7, Six1/2 and Six4/5 were expressed at high levels during the pediveliger stage. Moreover, Six1/2 and Six4/5 were highly expressed in mantle tissues. Subsequent overall in situ hybridization experiments in the planktonic larval stage revealed that Pax6, Six1/2 and Six4/5 detected signals in the region of the eyespot. Based on these analyses, we suggested that the development of vision in M. coruscus not only depended on the expression pattern of Pax6, but perhaps also related to Six1/2 and Six4/5 in the planktonic larval stage, while Six1/2 and Six4/5 were the dominant genes for visual function in the adult mussel. This study made a comprehensive analysis of the visual function of M. coruscus at the genome level, which helps us to understand the intrinsic mechanism of the visual system of marine bivalves, and also provides a molecular basis for improving the attachment and metamorphosis rate of M. coruscus larvae.

Compartmentalized pooling generates orientation selectivity in wide-field amacrine cells

Orientation is one of the most salient features in visual scenes. Neurons at multiple levels of the visual system detect orientation, but in many cases, the underlying biophysical mechanisms remain unresolved. Here, we studied mechanisms for orientation detection at the earliest stage in the visual system, in B/K wide-field amacrine cells (B/K WACs), a group of giant, nonspiking interneurons in the mouse retina that coexpress Bhlhe22 (B) and Kappa Opioid Receptor (K). B/K WACs exhibit orientation-tuned calcium signals along their long, straight, unbranching dendrites, which contain both synaptic inputs and outputs. Simultaneous dendritic calcium imaging and somatic voltage recordings reveal that individual B/K dendrites are electrotonically isolated, exhibiting a spatially confined yet extended receptive field along the dendrite, which we term "compartmentalized pooling." Further, the receptive field of a B/K WAC dendrite exhibits center-surround antagonism. Phenomenological receptive field models demonstrate that compartmentalized pooling generates orientation selectivity, and center-surround antagonism shapes band-pass spatial frequency tuning. At the microcircuit level, B/K WACs receive excitation driven by one contrast polarity (e.g., "ON") and glycinergic inhibition driven by the opposite polarity (e.g., "OFF"). However, this "crossover" inhibition is not essential for generating orientation selectivity. A minimal biophysical model reproduced compartmentalized pooling from feedforward excitatory inputs combined with a substantial increase in the specific membrane resistance between somatic and dendritic compartments. Collectively, our results reveal the biophysical mechanism for generating orientation selectivity in dendrites of B/K WACs, enriching our understanding of the diverse strategies employed throughout the visual system to detect orientation.

Visual perceptual artefacts in cardiothoracic imaging: Implications for diagnostic accuracy and strategies for mitigation.

Visual perceptual artefacts are distortions or illusions in medical image interpretation arising from the human visual system rather than hardware or imaging acquisition processes. These artefacts, emerging at various visual processing stages, such as the retina, visual pathways, visual cortex, and cognitive interpretation stages, impact the interpretation of cardiothoracic images. This review discusses artefacts including Mach bands, Dark Rim, Background Effects, Ambiguous Figures, Subjective Contours, and the Parallax Effect. A thorough literature search was conducted using PubMed and Google Scholar. Search terms included 'visual perceptual artifacts', 'cardiothoracic imaging', 'Mach bands', 'dark rim artifact', 'stress cardiac MRI', and 'radiology visual illusions'. Inclusion criteria encompassed studies focusing on visual perceptual artefacts in cardiothoracic imaging published in peer-reviewed journals. Studies not addressing visual perceptual artefacts or those focusing on imaging technology, hardware, or software-related artefacts were excluded. Duplicate studies were removed, and a final selection of 32 studies was analysed. Several key visual perceptual artefacts relevant to cardiothoracic imaging were identified. Mach bands, caused by lateral inhibition in the retina, create false contrasts at object boundaries, potentially mimicking pathologies such as fractures or pneumothoraces. The dark rim artefact observed on stress cardiac MRI, resulting from Gibbs ringing or truncation artefacts, can simulate perfusion defects and complicate diagnosis. Lateral inhibition can possibly accentuate the appearance of the dark rim artefact. Artefacts also fall under illusions of sensation, perception, and image formation. These illusions present both risks and benefits to diagnostic radiology. Recognizing visual perceptual artefacts is crucial for improving diagnostic accuracy and developing strategies to mitigate their impact. A comprehensive understanding and training on these artefacts are essential for radiologists to enhance interpretive skills, reduce diagnostic errors, and ultimately improve patient care in radiology.

Primary Visual Pathway Changes in Individuals With Chronic Mild Traumatic Brain Injury

Individuals with mild traumatic brain injury (TBI) often report vision problems despite having normal visual acuity and fundus examinations. Diagnostics are needed for these patients. To determine if a battery of assessments or machine-learning approaches can aid in diagnosing visual dysfunction in patients with mild TBI. This prospective, observational, case-control study was conducted between May 2018 and November 2021. The study setting was at a level 1 trauma research hospital. Participant eligibility included adult males and females with recorded best-corrected visual acuity and normal fundus examination. Individuals in the case group had a history of mild TBI; controls had no history of TBI. Exclusion criteria included a history of ocular, neurological, or psychiatric disease, moderate-severe TBI, recent TBI, metal implants, age younger than 18 years, and pregnancy. Cases and controls were sex- and age-matched. Data analysis was performed from July 2023 to March 2024. History of mild TBI in the case group. The single-session visit included the Neurobehavioral Symptom Inventory and measurements of oculomotor function, optical coherence tomography, contrast sensitivity, visual evoked potentials, visual field testing, and magnetic resonance imaging. A total of 28 participants (mean [SD] age, 35.0 [12.8] years; 15 male [53.6%]) with mild TBI and 28 controls (mean [SD] age, 35.8 [8.5] years; 19 female [67.9%]) were analyzed. Participants with mild TBI showed reduced prism convergence test breakpoint (-8.38; 95% CI, -14.14 to -2.62; P = .008) and recovery point (-8.44; 95% CI, -13.82 to -3.06; P = .004). Participants with mild TBI also had decreased contrast sensitivity (-0.07; 95% CI, -0.13 to -0.01; P = .04) and increased visual evoked potential binocular summation index (0.32; 95% CI, 0.02-0.63; P = .02). A subset of participants exhibited reduced peripapillary retinal nerve fiber layer thickness, increased optic nerve/sheath size, and brain cortical volumes. Machine learning identified subtle differences across the primary visual pathway, including the optic radiations and occipital lobe regions, independent of visual symptoms. Results of this case-control study suggest that the visual system was affected in individuals with mild TBI, even in those who did not self-report vision problems. These findings support the utility of a battery of assessments or machine-learning approaches to accurately diagnose this population.

Study of the double closed-loop active disturbance rejection control strategy for the longitudinal motion of fully submerged hydrofoil craft

In the process of high-speed driving, the longitudinal motion of the hydrofoil craft has the characteristics of parameter uncertainty and strong coupling, which leads to poor stability of the hydrofoil craft and the problem of high precision requirement for disturbance wave data. Therefore, a control method based on active disturbance rejection control (ADRC) and double closed loop is proposed. The algorithm is under the condition of decoupling the ship’s motion attitude and realizes the internal and external double-loop active disturbance rejection control of pitch angle and heave displacement based on the high-order sliding mode observer (HSMO) with better adaptability. By designing the speed error integral sliding mode surface, the dynamic characteristics of the system have been improved., and improve overall hydrofoil stability at high speeds. Finally, based on the Unreal Engine 5 platform, a visual longitudinal motion control system of hydrofoil craft is constructed. The simulation results show that compared to the existing sliding control mode, this control method can reduce the heave displacement and pitch angle by about 50%, shorten the response time of real-time control of hydrofoil craft. The superiority and effectiveness of the control system were verified.

Advances in Robotic Surgery: A Review of New Surgical Platforms

In recent decades, the development of surgical systems which minimize patient impact has been a major focus for surgeons and researchers, leading to the advent of robotic systems for minimally invasive surgery. These technologies offer significant patient benefits, including enhanced outcome quality and accuracy, reduced invasiveness, lower blood loss, decreased postoperative pain, diminished infection risk, and shorter hospitalization and recovery times. Surgeons benefit from the elimination of human tremor, ergonomic advantages, improved vision systems, better access to challenging anatomical areas, and magnified 3DHD visualization of the operating field. Since 2000, Intuitive Surgical has developed multiple generations of master-slave multi-arm robots, securing over 7000 patents, which created significant barriers for competitors. This monopoly resulted in the widespread adoption of their technology, now used in over 11 million surgeries globally. With the expiration of key patents, new robotic platforms featuring innovative designs, such as modular systems, are emerging. This review examines advancements in robotic surgery within the fields of general, urological, and gynecological surgery. The objective is to analyze the current robotic surgical platforms, their technological progress, and their impact on surgical practices. By examining these platforms, this review provides insights into their development, potential benefits, and future directions in robotic-assisted surgery.

Short-Wave Infrared (SWIR) Imaging for Robust Material Classification: Overcoming Limitations of Visible Spectrum Data

This paper presents a novel approach to material classification using short-wave infrared (SWIR) imaging, aimed at applications where differentiating visually similar objects based on material properties is essential, such as in autonomous driving. Traditional vision systems, relying on visible spectrum imaging, struggle to distinguish between objects with similar appearances but different material compositions. Our method leverages SWIR’s distinct reflectance characteristics, particularly for materials containing moisture, and demonstrates a significant improvement in accuracy. Specifically, SWIR data achieved near-perfect classification results with an accuracy of 99% for distinguishing real from artificial objects, compared to 77% with visible spectrum data. In object detection tasks, our SWIR-based model achieved a mean average precision (mAP) of 0.98 for human detection and up to 1.00 for other objects, demonstrating its robustness in reducing false detections. This study underscores SWIR’s potential to enhance object recognition and reduce ambiguity in complex environments, offering a valuable contribution to material-based object recognition in autonomous driving, manufacturing, and beyond.

Image-based weight estimation and grading of post-harvested Indian potato (Solanum tuberosum) using dimensional analysis integrated with adaptive neuro fuzzy inference system approach

ABSTRACT The current study builds a model of the link between the image-based weights of Indian Potatoes extracted using the built computer vision system using a dimensional analysis integrated with the adaptive neuro-fuzzy inference system. An atomized sorting system must be developed to enhance the exportation quality. A soft computing method is applied to nonlinear data training. The features include size (dimensions, area, and volumes), shapes, and gravimetric properties were chosen. As per standards and export criteria, potatoes should be divided into three groups (small, medium, and big) based on their estimated weight. A self-build image-based system was designed for capturing an image. The experimental outcomes were contrasted with the presented integrated predicted outcomes. A respectable level of agreement is noted with good competency and dependability by the strong correlation coefficient (0.9975). Hence, the presented approach is strongly recommended for image-based weight estimation and classification. Code: https://colab.research.google.com/drive/1g-B81q8284Hw1QFwAJjqpmDEV35KsNzV

Enhancing Real-Time Vision Systems: Integrating Dynamic Vision Sensors with Vision Transformers to Increase Computational Efficiency

Abstract. Optimizing and understanding vision systems' computational accuracy is crucial as they become increasingly integrated into daily life. Dynamic Vision Sensors (DVS) and Vision Transformers (ViTs) lead computer vision technology with efficient object recognition and image processing. However, DVS data's high computational complexity poses a problem for its real-time implementations. Merging these technologies can enhance vision system performance in dynamic environments and optimize real-time DVS processing. In our work, we use a ViT architecture to classify the DVS 128 dataset and compare our results with existing works using SNNs. We analyze how our method affects accuracy and loss, experimenting with different DVS-to-ViT input patch sizes. Our results show that the large patch size of 32x32 pixels has better accuracy and smaller loss than the 4x4 pixel patches as epochs increase. Our method also achieved a high 98.4% accuracy and low 0.22 loss within five epochs, significantly outperforming previous works averaging 93.13% accuracy over more epochs. These results highlight ViTs' large potential for real-time DVS data classification in applications that require high accuracy, like autonomous vehicles and surveillance systems.

Shadow Removal for Enhanced Nighttime Driving Scene Generation

Autonomous vehicles depend on robust vision systems capable of performing under diverse lighting conditions, yet existing models often exhibit substantial performance degradation when applied to nighttime scenarios after being trained exclusively on daytime data. This discrepancy arises from the lack of fine-grained details that characterize nighttime environments, such as shadows and varying light intensities. To address this gap, we introduce a targeted approach to shadow removal designed for driving scenes. By applying Partitioned Shadow Removal, an enhanced technique that refines shadow-affected areas, alongside image-to-image translation, we generate realistic nighttime scenes from daytime data. Experimental results indicate that our augmented nighttime scenes significantly enhance segmentation accuracy in shadow-impacted regions, thereby increasing model robustness under low-light conditions. Our findings highlight the value of Partitioned Shadow Removal as a practical data augmentation tool, adapted to address the unique challenges of applying shadow removal in driving scenes, thereby paving the way for improved nighttime performance in autonomous vehicle vision systems.

Validation of a Novel Strategy for Fluorescence Quenching for a Self-Quenching Fluorogenic Probe and Its Application for Visual Loop-Mediated Isothermal Amplification Detection During Food Safety Analysis

The self-quenching fluorogenic probe facilitates precise identification of LAMP (loop-mediated isothermal amplification) amplicons, unaffected by non-specific products resulting from primer dimers. However, low quenching efficiency by surrounding nucleobases leads to high background signal, posing significant challenges for visual inspection with the naked eye. The present study aims to identify an oligonucleotide sequence that is complementary to the self-quenching fluorogenic probe, and to employ the fluorescence super-quenching mechanism of double-stranded DNA to establish a visualization system for the LAMP assay. The results indicated that the incorporation of a sequence fully complementary to the probe could significantly reduce the system’s background fluorescence (p < 0.05). When the melting temperature exceeds room temperature, truncating the complementary sequence from the 3′ end does not compromise the probe’s quenching efficiency. The LAMP visualization system, using a 10–13-base complementary sequence of the loop primer-based probe, could effectively minimize background fluorescence and yield straightforward visual results post-reaction. Applied to rainbow trout and Atlantic salmon detection, the system detected 1 pg DNA in a closed-tube format. In conclusion, a suitable complementary sequence can reduce the background fluorescence of the self-quenching fluorogenic probe. Employing this sequence alongside the self-quenching fluorogenic probe to develop a low-background fluorescence LAMP system demonstrates great potential for successful visual detection and holds considerable promotional merit.

Recent Advances in Visual SLAM Algorithms in Dynamic Environments

This paper reviews the latest research progress of visual SLAM in dynamic scenes. The improved algorithms dependent on deep learning and multi-sensor fusion are discussed. First of all, the challenges of traditional visual SLAM in dynamic environments are stated, especially the interference of high dynamic objects on system feature extraction, pose estimation and mapping. Then, the application of several object detection technologies such as YOLOv5 and YOLOv7 in dynamic visual SLAM was analyzed in detail. These methods significantly improve the positioning precision and robustness of the system by identifying and eliminating dynamic feature points in the scene. In addition, this paper explores the visual SLAM system combining semantic segmentation and geometric constraint optimization, and finally introduces the multi-sensor SLAM method combining IMU data, point cloud data and visual information. Experimental results show that these improved algorithms have a significant improvement in positioning accuracy and system robustness compared with traditional visual SLAM algorithms under TUM and other dynamic data sets.

Advancing Forest Plot Surveys: A Comparative Study of Visual vs. LiDAR SLAM Technologies

Forest plot surveys are vital for monitoring forest resource growth, contributing to their sustainable development. The accuracy and efficiency of these surveys are paramount, making technological advancements such as Simultaneous Localization and Mapping (SLAM) crucial. This study investigates the application of SLAM technology, utilizing LiDAR (Light Detection and Ranging) and monocular cameras, to enhance forestry plot surveys. Conducted in three 32 × 32 m plots within the Tibet Autonomous Region of China, the research compares the efficacy of LiDAR-based and visual SLAM algorithms in estimating tree parameters such as diameter at breast height (DBH), tree height, and position, alongside their adaptability to forest environments. The findings revealed that both types of algorithms achieved high precision in DBH estimation, with LiDAR SLAM presenting a root mean square error (RMSE) range of 1.4 to 1.96 cm and visual SLAM showing a slightly higher precision, with an RMSE of 0.72 to 0.85 cm. In terms of tree position accuracy, the three methods can achieve tree location measurements. LiDAR SLAM accurately represents the relative positions of trees, while the traditional and visual SLAM systems exhibit slight positional offsets for individual trees. However, discrepancies arose in tree height estimation accuracy, where visual SLAM exhibited a bias range from −0.55 to 0.19 m and an RMSE of 1.36 to 2.34 m, while LiDAR SLAM had a broader bias range and higher RMSE, especially for trees over 25 m, attributed to scanning angle limitations and branch occlusion. Moreover, the study highlights the comprehensive point cloud data generated by LiDAR SLAM, useful for calculating extensive tree parameters such as volume and carbon storage and Tree Information Modeling (TIM) through digital twin technology. In contrast, the sparser data from visual SLAM limits its use to basic parameter estimation. These insights underscore the effectiveness and precision of SLAM-based approaches in forestry plot surveys while also indicating distinct advantages and suitability of each method to different forest environments. The findings advocate for tailored survey strategies, aligning with specific forest conditions and requirements, enhancing the application of SLAM technology in forestry management and conservation efforts.

The Interconnectedness Of Cognitive And Formative Skills In Elementary Students With Specific Reading Disabilities: Insights For Early Education

The current research aims to investigate the development of formative abilities in elementary school-aged children with specific reading disabilities and its relationship to reading efficiency. In this context, the analysis depicted poor performance in design tasks exhibited by SRD students compared to regular readers and GLD students. The researchers attribute these results to cognitive failures within the concerned mechanisms, including orientation, sequence, and organization of the Visual Analysis System. The findings indicate that these deficiencies affect not only visual representation but also reading and writing. More importantly, there were solid correlations between design abilities, perceptual-motor skills, and memory, pointing to the interrelationship between those skills. As a result, the findings recommend highly targeted interventions at an early stage of education and suggest the need to incorporate creative design-based activities into reading skill enhancement and support for students with special needs.

A Lightweight Deep-Learning Visual SLAM for Indoor Dynamic Environment Using Yolov10

Vision simultaneous localization and mapping (SLAM) technology has become a key research direction in the field of mobile robotics in recent years. However, the accuracy and stability of traditional vision SLAM technology greatly affected by the dynamic environment, and the mainstream dynamic feature point rejection method combining vision and semantic segmentation techniques is not applicable to edge-end devices with limited resources and high real-time requirements. This research suggests a visual SLAM algorithm based on the YOLOv10n lightweight target detection model and the GCNv2 feature point extraction model to accomplish real-time dynamic feature point rejection to address those problems. To compensate for the detection accuracy and stability issues of the YOLOv10n model while leveraging its real-time advantages, the algorithm also employs multi-target Kalman filtering with data association through the Hungarian algorithm, history window smoothing, and potential dynamic feature point recording methods to enhance robustness. The algorithm is validated on the TUM RGB-D Dataset, and the results demonstrate that the method can effectively reject the dynamic feature points in the dynamic environment, and it has a significant improvement in the accuracy and stability of the visual SLAM system.