High-definition Images Research Articles

A novel TMTP1-modified theranostic nanoplatform for targeted in vivo NIR-II fluorescence imaging-guided chemotherapy for cervical cancer.

Near-infrared II (NIR-II, 900-1700 nm) fluorescence bioimaging with advantages of good biosafety, excellent spatial resolution, high sensitivity, and contrast has attracted great attention in biomedical research fields. However, most of the nanoprobes used for NIR-II fluorescence imaging have poor tumor-targeting ability and therapeutic efficiency. To overcome these limitations, a novel NIR-II-emissive theranostic nanoplatform for fluorescence imaging and treatment of cervical cancer was designed and prepared. The NIR-II-emissive dye IR-783 and chemotherapy drug doxorubicin (DOX) were encapsulated into liposomes, and the tumor-targeting peptide TMTP1 (a polypeptide with a sequence of cyclic ASN Val Val Arg Gln Cys) was conjugated to the surface of the liposomes to form IR-783-DOX-TMTP1 nanoparticles (NPs) via self-assembly methods. The IR-783-DOX-TMTP1 NPs showed strong NIR-II emission, excellent biocompatibility and a long lifetime in vivo. Furthermore, high-definition NIR-II fluorescence microscopy images of ear blood vessels and intratumoral blood vessels were obtained from IR-783-DOX-TMTP1 NP-stained mice with high spatial resolution under 808 nm laser excitation. Moreover, IR-783-DOX-TMTP1 NPs showed strong tumor-targeting ability and highly efficient chemotherapeutic characteristics towards cervical tumors. The novel targeting and NIR-II-emissive IR-783-DOX-TMTP1 NPs have great potential in diagnosis and therapy for cervical cancer.

Demonstration experiment of telemedicine using ultrasonography and telerehabilitation with 5G communication system in aging and depopulated mountainous area.

ObjectiveThe challenges of an aging population worldwide are the increased number of people needing medical and nursing care and inadequate medical resources. Information and communication technologies have progressed remarkably, leading to innovations in various areas. 5G communication systems are capable of high-capacity, high-speed communication with low latency and are expected to transform medicine. We aimed to report a demonstration experiment of telerehabilitation and telemedicine using a mobile ultrasound system in a depopulated area in a mountainous terrain, where 32% of the population are 65 years or older.MethodsAt the core hospital, a physician or physical therapist remotely performed ultrasonography or rehabilitation on a subject in a clinic. Five general residents participated in the telerehabilitation as subjects. The delay time and video quality transmitted with 5G and long-term evolution (LTE) communication systems were compared. The physician or physical therapist subjectively evaluated the quality and delay of the transmitted images and subject acceptability.ResultsOf seven physical therapists, six and three responded that the video quality was “good” for telerehabilitation with 5G/4K resolution and LTE, respectively. Five physical therapists and one physical therapist reported that the delay time was “acceptable” with 5G/4K resolution and LTE, respectively. For telemedicine using a mobile ultrasound system, the responses for 5G were “the delay was acceptable” and “rather acceptable.” In contrast, both respondents’ responses for LTE were “not acceptable.”ConclusionsMultiple high-definition images can be transmitted with lower latency in telerehabilitation and telemedicine using mobile ultrasound imaging systems with a 5G communication system. These differences affected the subjective evaluation of the doctors and physical therapists.

2D Multi-view Discriminant Analysis

With the advent of technology, we have seen an explosion in high-definition images. However, due to the large size of images, 1D methods are less suitable to handle such data. This issue becomes even more pronounced when working with multi-view datasets. We propose a generalized method, 2D Multi-view Discriminant Analysis (2DMvDA), to tackle this issue. 2DMvDA is a 2D multi-view classification method based on discriminant analysis. It uses the 2D image matrices directly instead of extracting 1D features from them, leading to a considerable reduction in the size of the data to be processed. We compare the proposed method with 1D Multi-view Discriminant Analysis (MvDA) and single-view 2D Linear Discriminant Analysis (2DLDA). We experimentally show that the proposed method requires less than 0.1% training time and memory than MvDA. Also, though it needs the same memory, it requires 30–70% less training time than 2DLDA. Using less time and memory, 2DMvDA achieves a classification accuracy that is at par or better than both of these methods.

Incremental benefit of dye-based chromoendoscopy to predict the risk of submucosal invasive cancer in large nonpedunculated colorectal polyps

Detailed lesion assessment of large nonpedunculated colorectal polyps (LNPCPs;≥20mm) can help predict the risk of submucosal invasive cancer (SMIC). Traditionally this has required the use of dye-based chromoendoscopy (DBC). We sought to assess the accuracy and incremental benefit of DBC in addition to high-definition white-light imaging (HDWLI) and virtual chromoendoscopy (VCE) for the prediction of SMIC within LNPCPs. A prospective observational study of consecutive LNPCPs at a single tertiary referral center was performed. Before resection all lesions were assessed for the presence of a demarcated area (DA), defined as an area of disordered pit or microvascular pattern, by 2 trained endoscopists before and after DBC. Diagnostic performance characteristics were calculated with histology as the reference criterion standard, and overall agreement was calculated using the κ statistic. Over 39 months to March 2021, 400 consecutive LNPCPs (median lesion size, 35mm; interquartile range, 25-45) were analyzed. The overall rate of SMIC was 6.5%. Presence of a DA had an accuracy of 91% (95% confidence interval, 87.7-93.5) for SMIC, independent of the use of DBC. The rate of interobserver agreement for presence of a DA using HDWLI+ VCE was very high (κ= .96) with no benefit gained by the addition of DBC. The use of HDWLI and VCE is likely to be adequate for lesion assessment for the prediction of SMIC among LNPCPs. Further, the absence of a DA is strongly predictive for the absence of SMIC, independent to the use of DBC. (Clinical trial registration number: NCT03506321.).

Does intra-articular injection of adipose-derived stem cells improve cartilage mass? A case report using three-dimensional image analysis software in knee osteoarthritis

BackgroundMesenchymal stem cells are currently a research focus because of the possibility of cartilage regeneration through several mechanisms, including mesenchymal stem cell sheets. However, there are no published reports visualizing cartilage in three dimensions. Here, we report a case of improved cartilage volume. We purified and cultured adipose-derived mesenchymal stem cells and then performed adipose-derived mesenchymal stem cell therapy by directly injecting these cells into the articular cartilage. Cartilage was quantitatively evaluated before and after injection using three-dimensional image analysis software based on the magnetic resonance imaging.Case presentationThe patient, a 55-year-old Japanese woman, experienced pain in both knees and was diagnosed with osteoarthritis of the knee. We performed adipose-derived mesenchymal stem cell therapy in both knees at our hospital and quantitatively evaluated cartilage before and after the treatment using the three-dimensional image analysis software “SYNAPSE VINCENT”.ConclusionsPreoperatively, the cartilage defect area was 33.59 mm2 in the femur and 122.31 mm2 in the tibia; however, 12 months postoperatively, it improved to 13.59 mm2 and 51.43 mm2, respectively. Furthermore, the preoperative femur and tibia volumes were 9.58 mL and 3.82 mL, respectively; however, 12 months postoperatively, these values improved to 10.00 mL and 4.17 mL, respectively. For the quantitative analysis of cartilage, SYNAPSE VINCENT visualizes the state of cartilage in a high-definition three-dimensional image, which is excellent for understanding the state of the disease and explaining it to the patient. Although SYNAPSE VINCENT can only analyze the thickness of cartilage, and the reproducibility of the error is debatable, SYNAPSE VINCENT would be useful as a clinical tool for regenerative medicine. We have shown in this case report the promising effects of adipose-derived stem cell intraarticular injections in treating osteoarthritis and the use of new diagnostic instruments.

Quantitative Biomarkers for Cancer Detection Using Contrast-Free Ultrasound High-Definition Microvessel Imaging: Fractal Dimension, Murray's Deviation, Bifurcation Angle & Spatial Vascularity Pattern.

A growing body of evidence indicates that there is a strong correlation between microvascular morphological features and malignant tumors. Therefore, quantification of these features might allow more accurate differentiation of benign and malignant tumors. The main objective of this research project is to improve the quantification of microvascular networks depicted in contrast-free ultrasound microvessel images. To achieve this goal, a new series of quantitative microvessel morphological parameters are introduced for differentiation of breast masses using contrast-free ultrasound-based high-definition microvessel imaging (HDMI). Using HDMI, we quantified and analyzed four new parameters: 1) microvessel fractal dimension (mvFD), a marker of tumor microvascularcomplexity; 2) Murray’sdeviation(MD), the diameter mismatch, defined as the deviation from Murray’s law; 3) bifurcation angle (BA), abnormally decreased angle; and 4) spatial vascular pattern (SVP), indicating tumor vascular distribution pattern, either intratumoral or peritumoral. The new biomarkers have been tested on 60 patients with breast masses. Validation of the feature’s extraction algorithm was performed using a synthetic data set. All the proposed parameters had the power to discriminate the breast lesion malignancy (p < 0.05), displaying BA as the most sensitive test, with a sensitivity of 90.6%, and mvFD as the most specifictest, with a specificityof 92%. The results of all four new biomarkers showed an AUC = 0.889, sensitivity of 80% and specificity of 91.4% In conclusion, the added value of the proposed quantitative morphological parameters, as new biomarkers of angiogenesis within breast masses, paves the way for more accurate breast cancer detection with higher specificity.

Research on Structural Analysis Method of Long-span Steel Structure Construction Process Based on Feature Extraction Algorithm

The construction of long-span steel structure is a continuous process, and the stress state of its structure changes. Modern construction projects tend to be more and more high-rise buildings, because the span and height of construction projects are small at the early stage of development of construction industry, and the stress of building structures is not large compared with high-rise and large-span buildings, so the situation of components after construction is roughly the same as that before construction. Image mosaic technology is an important branch of image processing technology. Its significance is to use small image acquisition equipment to obtain large and high-definition images through software splicing. On the one hand, more and more complete information can be obtained on an image through splicing. On the other hand, the cost of small equipment is much lower than that of large equipment, which can greatly reduce the cost. Combined with feature extraction algorithm, this paper expounds the structural analysis method of long-span steel structure construction process.

Optimization and regularization of complex task decomposition for blind removal of multi-factor degradation

Most existing image restoration methods based on deep neural networks are developed for images which only degraded by a single degradation mode and imaging under an ideal condition. They cannot be directly used to restore the images degraded by multi-factor coupling. A complex task decomposition regularization optimization strategy (TDROS) is proposed to solve the problem. The restoration of images degraded by multi-factor coupling is a complex task that can be solved by separating these multiple factors, that is, breaking the complex task into numbers of simpler tasks to make the entire complex problem be overcome more easily. Motivated by this idea, the TDROS decomposes the complex task of image restoration into two sub-task: the potential task constrained by regularization and the main task for reconstructing high-definition images. In TDROS, the front of the neural network is focused on the restoration of images degraded by additive noise, while the other part of the network is focused mainly on the restoration of images degraded by blur. We applied the TDROS to an 11-layer convolutional neural network (CNN) and compared it with initial CNNs from the aspects of restoration accuracy and generalization ability. Based on these results, we used TDROS to design a novel network model for the restoration of atmospheric turbulence-degraded images. The experimental results demonstrate that the proposed TDROS can improve the generalization ability of the existing network more effectively than current popular methods, offering a better solution for the problem of severely degraded image restoration. Moreover, the TDROS concept provides a flexible framework for low-level visual complex tasks and can be easily incorporated into existing CNNs.

A Practical Method for the Automatic Recognition of Rock Structures in Panoramic Borehole Image during Deep-Hole Drilling Engineering

Digital panoramic borehole imaging technology has been widely used in the practice of drilling engineering. Based on many high-definition panoramic borehole images obtained by the borehole imaging system, this paper puts forward an automatic recognition method based on clustering and characteristic functions to perform intelligent analysis and automatic interpretation researches, and successfully applied to the analysis of the borehole images obtained at the Wudongde Hydropower Station in the south-west of China. The results show that the automatic recognition method can fully and quickly automatically identify most of the important structural planes and their position, dip, dip angle and gap width and other characteristic parameter information in the entire borehole image. The recognition rate of the main structural plane is about 90%. The accuracy rate is about 85%, the total time cost is about 3 h, and the accuracy deviation is less than 4% among the 12 boreholes with a depth of about 50 m. The application of automatic recognition technology to the panoramic borehole image can greatly improve work efficiency, reduce the time cost, and avoid the interference caused by humans, making it possible to automatically recognize the structural plane parameters of the full-hole image.

Dual-input convolutional neural network for glaucoma diagnosis using spectral-domain optical coherence tomography

Background/AimsTo evaluate, with spectral-domain optical coherence tomography (SD-OCT), the glaucoma-diagnostic ability of a deep-learning classifier.MethodsA total of 777 Cirrus high-definition SD-OCT image sets of the retinal nerve fibre layer (RNFL)...

Robust location-secured high-definition image watermarking based on key-point detection and deep learning

In this paper, we propose a blind high-definition image watermarking scheme named multiscale-fusion dilated ResNet-based high-definition watermarking (MDResNet-HDWM), which is based on a key-point detection and deep learning framework. The proposed watermarking scheme embeds watermark in multiple non-overlapping regions, whose locations are secured with a private key referring to several dominant key points. To achieve scale-invariant watermark embedding, regions are first determined in a normalized image copy and then mapped back to its origin. The watermarks are embedded in the central part of the region in such a way that the watermarks will always be located inside the regions even if the image is geometrically transformed. The watermarks are embedded and extracted with MDResNet, which is trained with a curriculum learning strategy that makes it be robust to signal processing operations and geometric transforms. Experimental results demonstrate that the proposed MDResNet-HDWM achieves good performance and is robust to both common signal operations and geometric attacks.

Colour me blue: chromoendoscopy and advanced detection methods in ulcerative colitis.

Surveillance colonoscopy is recommended in patients with long standing ulcerative colitis or Crohn's colitis. Chromoendoscopy is the main technique for increased detection of colitis-associated dysplasia. However, the recommendation was made on the basis of studies using standard definition colonoscopes. This review highlights randomized controlled trials and meta-analysis, which were published between 2018 and 2021 with the focus of conventional chromoendoscopy, virtual chromoendoscopy and high-definition imaging. In addition, studies investigating the value of random biopsies were also evaluated. Chromoendoscopy increases the total number of colitis-associated dysplasia even by using high-definition colonoscopes. However, the procedure time is prolonged and there is no significant difference in the diagnostic yield of high definition alone and high definition with chromoendoscopy. Virtual chromoendoscopy seems not to develop a role for surveillance in inflammatory bowel disease (IBD) patients. High-definition colonoscopy and conventional chromoendoscopy are key techniques for surveillance in IBD.

Deep reinforcement learning-based multi-objective control of hybrid power system combined with road recognition under time-varying environment

Aiming at promoting the intelligent development of control technology for new energy vehicles and showing the outstanding advantages of deep reinforcement learning (DRL), this paper trained a VGG16-based road recognition convolutional neural network (CNN) at first. Lots of high-definition images of five typical roads are collected by the racing game Dust Rally 2.0, including dry asphalt, wet asphalt, snow, dry cobblestone, and wet cobblestone. Then, a time-varying driving environment model was established, involving driving images, road slope, longitudinal speed, and the number of passengers. Finally, a stereoscopic control network suitable for nine-dimensional state space and three-dimensional action space was built, and for parallel hybrid electric vehicles (HEVs) with the P3 structure, a deep q-network (DQN)-based energy management strategy (EMS) achieving multi-objective control was proposed, including the fine-tuning strategy of the motor speed to maintain the optimal slip rate during braking, the engine power control strategy and the continuously variable transmission (CVT) gear ratio control strategy. Simulation results show under the influence of some factors such as tree shade and image compression, the road recognition network has the highest accuracy for snow roads and wet asphalt roads. Three types of control strategies learned simultaneously by the stereoscopic control network not only maintain the near-optimal slip rate in the braking period but also achieve a fuel consumption of 4788.93 g, while dynamic programming (DP)-based EMS gets a fuel consumption of 4295.61 g. Moreover, even DP-based EMS only contains three states and two actions, the time consumed for DP-based EMS and DQN-based EMS to run the speed cycle of 3602s is about 4911s and 10s, respectively. Therefore, the optimization and real-time performance of DRL-based EMS can be guaranteed.

Data gap decomposed by auxiliary modality for NIR‐VIS heterogeneous face recognition

In the dark scene at night, the face images captured by ordinary visible light (VIS) are generally poor quality and very dim, while the near-infrared (NIR) can capture high definition and recognizable face images at night. The NIR-VIS Heterogeneous face recognition has become a hot research field, which helps to build an all-weather face recognition system. NIR-VIS HFR is sophisticated because of the large visual difference between NIR images and VIS images. In order to reduce the difficulty of such cross-modality invariant feature learning, this paper proposes a cross-modality data gap decomposed by auxiliary modality method (DGD) for NIR-VIS HFR. First, the brightness component (Y component) of VIS image YCbCr space is used as the auxiliary modality to decompose the cross-modality data gap. The lightness component retained the structural information of VIS image and was similar to the colour information of NIR modality; in this way, the huge gap between the NIR data and the VIS data is decomposed into two smaller gaps, thus reducing the difficulty of network learning. Second, the data of the three modalities are input into the weight sharing network and training under the combined guidance of cross-modality gap decomposition loss and intra-modality gap loss; in this way, the modality invariant features can be learned faster and better. Extensive experiments were conducted on two commonly used datasets CASIA NIR-VIS 2.0 and Oulu-CASIA NIR-VIS to evaluate DGD method. Experimental results indicate DGD method has competitive performance compared with the latest methods.

Performance enhancement method for multiple license plate recognition in challenging environments

Multiple-license plate recognition is gaining popularity in the Intelligent Transport System (ITS) applications for security monitoring and surveillance. Advancements in acquisition devices have increased the availability of high definition (HD) images, which can capture images of multiple vehicles. Since license plate (LP) occupies a relatively small portion of an image, therefore, detection of LP in an image is considered a challenging task. Moreover, the overall performance deteriorates when the aforementioned factor combines with varying illumination conditions, such as night, dusk, and rainy. As it is difficult to locate a small object in an entire image, this paper proposes a two-step approach for plate localization in challenging conditions. In the first step, the Faster-Region-based Convolutional Neural Network algorithm (Faster R-CNN) is used to detect all the vehicles in an image, which results in scaled information to locate plates. In the second step, morphological operations are employed to reduce non-plate regions. Meanwhile, geometric properties are used to localize plates in the HSI color space. This approach increases accuracy and reduces processing time. For character recognition, the look-up table (LUT) classifier using adaptive boosting with modified census transform (MCT) as a feature extractor is used. Both proposed plate detection and character recognition methods have significantly outperformed conventional approaches in terms of precision and recall for multiple plate recognition.

Research on the Application of UAV Wide-angle Shooting Technology in Multimedia Video Playing

Unmanned aerial vehicles continue to increase the requirements for image resolution in the fields of reconnaissance, surveying, and mapping. The fusion of the video image system and the traditional remote control and telemetry system also needs to be considered in the design. Aiming at the above problems, the article designs and implements an airborne high-definition video image system based on FPGA+DSP. The system can store the video image on the local hard disk, and at the same time, the compressed video image is also convenient for real-time transmission to the ground control station through the wireless link equipment. The system detection found that the buffer occupancy can fluctuate smoothly in a small range, and the subjective image quality has also been improved, which better meets the low-latency application requirements of the airborne reconnaissance system.

Spatial sampling effect on data structure and random forest classification of tissue types in High Definition and Standard Definition FT-IR imaging

Infrared radiation imaging (IR) combined with machine learning algorithms is one of the most promising techniques in tissue structures recognition and cancer diagnosis. Pancreatic tissue is heterogeneous and provides variable biochemical composition, as well as exhibits a variety of shapes and sizes having strong impact on physical aspects of light absorption. This research explores the potential of High and Standard Definitions in pancreatic tissue type prediction with random forest classification based on different spectral ranges. IR images spatial resolution depends on wavenumber and objective's numerical aperture. Therefore, data measured in Standard Definition (SD) are affected by under-sampling in the high wavenumbers while High Definition (HD) is free of this limitation for the whole analyzed spectral region. In order to investigate this effect a Random Forest model was created using a set of 56 biopsies measured in both definitions. In terms of signal variance the HD was found to be more spread out (however, with smaller scattering). SD models obtained a higher True Positive Rate than HD, however, spatial detail was better in HD models. Models trained on one definition and predicted on the other suggest that HD model can be used to successfully predict SD data. Finally, models based on only fingerprint and only high wavenumber were created to assess information content in each region. Models based on fingerprint region had very good prediction results across all classes, while for high wavenumber region results were class and definition dependent.

Seabed features associated with cold seep activity at the Formosa Ridge, South China Sea: Integrated application of high-resolution acoustic data and photomosaic images

High resolution acoustic data, together with high-definition photomosaic images and video observations of the Formosa Ridge Cold Seep Site were obtained using a remotely operated vehicle (ROV). Seabed features of the cold seep site were described in detail and the scales and distribution patterns of these features displayed on the seabed were quantitatively analyzed. The cold seep site is characterized by irregular and hummocky topography on the seabed relief map and patches with high to medium backscatter intensities on side-scan sonar (SSS) images. Based on visual data, six kinds of seabed features related to cold seep activity were identified: gas flares, dense benthic chemosynthetic communities, mussel beds with low population densities, shells and/or shell debris areas, authigenic carbonate concretions, and reduced sediments. Spatially, these seabed features are distributed in an approximately zoned pattern with gas plumes and benthic communities in the innermost part of Site F and the reduced sediments in the outermost part. Based on the integrated acoustic and visual images and in-situ geochemical data, a conceptual model describing the cold seep fluid conduits within the shallow subsurface at Site F at the present is established. Briefly, favorable conduits have developed within the authigenic carbonate mound. The conduits channelize the cold seep fluids and result in relatively focused discharge of the seep fluids at the intersections of the conduits and the seabed, to sustain the dense communities there. Additionally, the authigenic carbonate concretions redirect some of the cold seep fluids to seeps at the periphery of the site, resulting in the reduced sediments. Integrated acoustic and visual investigations combined with in situ measurements are an effective way to quantitatively study the spatial extent of a cold seep site and can provide detailed information for further study of the cold seep system.

A Background Assessable and Correctable Bimolecular Fluorescence Complementation System for Nanoscopic Single-Molecule Imaging of Intracellular Protein-Protein Interactions.

Bimolecular Fluorescence Complementation (BiFC) is a versatile approach for intracellular analysis of protein-protein interactions (PPIs), but the tendency of the split fluorescent protein (FP) fragments to self-assemble when brought into close proximity of each other by random collision can lead to generation of false-positive signals that hamper high-definition imaging of PPIs occurring on the nanoscopic level. While it is thought that expressing the fusion proteins at a low level can remove false positives without impacting specific signals, there has been no effective strategy to test this possibility. Here, we present a system capable of assessing and removing BiFC false positives, termed Background Assessable and Correctable-BiFC (BAC-BiFC), in which one of the split FP fragments is fused with an optically distinct FP that serves as a reference marker, and the single-cell fluorescence ratio of the BiFC signal to the reference signal is used to gauge an optimal transfection condition. We showed that when BAC-BiFC is designed to image PPIs regulating Human Immunodeficiency Virus type 1 (HIV-1) assembly, the fluorescence ratio could decrease with decreasing probe quantity, and ratios approaching the limit of detection could allow physiologically relevant characterization of the assembly process on the nanoscale by single-molecule localization microscopy (SMLM). With much improved clarity, previously undescribed features of HIV-1 assembly were revealed.

Semantic segmentation model of cotton roots in-situ image based on attention mechanism

The growth and distribution of root system in the soil has an important influence on the growth of plants and is one of the important factors affecting crop production. However, the root system of plants is located in the dark and closed soil. Even if we can obtain high-definition root image from the complex soils, the interference of the soil particles on root system and the small difference of color between them will pose challenges for further root segmentation. In this experiment, the cotton mature root system is used as the research object. Based on the introduction of sub-pixel convolution DeepLabv3+ semantic segmentation model, we further added the attention mechanism to the model, assigning more weight to the pixel points of fine roots and their root hairs, and designed a semantic segmentation model of cotton roots in-situ image based on the attention mechanism. The experimental results show that the model has higher segmentation accuracy and operational efficiency than only introduces sub-pixel convolution DeepLabv3+ model, U-Net model and SegNet model. The precision value, recall value and F1-score are 0.9971, 0.9984 and 0.9937 respectively, and the IoU value of 161 untrained root image segmentation tasks was 0.9875. At the same time, we also performed segmentation experiments on the early cotton root images. The results show that the DeepLabv3+ model which only introduces sub-pixel convolution, U-Net model and SegNet model have poor segmentation effects. The semantic segmentation model based on attention mechanism proposed in this paper can be segmented accurately. The above results show that the proposed model can distinguish the cotton root system from the complex soil background accurately and has good segmentation effect. It can realize the accurate segmentation of root image in early and mature period in the process of cotton root growth, and provide important theoretical value and practical application reference for deep learning in plant root segmentation.