Deep learning has become popular and the mainstream technology in many researches related to learning,and has shown its impact on photogrammetry.According to the definition of photogrammetry,that is,a subject that res...Deep learning has become popular and the mainstream technology in many researches related to learning,and has shown its impact on photogrammetry.According to the definition of photogrammetry,that is,a subject that researches shapes,locations,sizes,characteristics and inter-relationships of real objects from optical images,photogrammetry considers two aspects,geometry and semantics.From the two aspects,we review the history of deep learning and discuss its current applications on photogrammetry,and forecast the future development of photogrammetry.In geometry,the deep convolutional neural network(CNN)has been widely applied in stereo matching,SLAM and 3D reconstruction,and has made some effects but needs more improvement.In semantics,conventional methods that have to design empirical and handcrafted features have failed to extract the semantic information accurately and failed to produce types of“semantic thematic map”as 4D productions(DEM,DOM,DLG,DRG)of photogrammetry.This causes the semantic part of photogrammetry be ignored for a long time.The powerful generalization capacity,ability to fit any functions and stability under types of situations of deep leaning is making the automatic production of thematic maps possible.We review the achievements that have been obtained in road network extraction,building detection and crop classification,etc.,and forecast that producing high-accuracy semantic thematic maps directly from optical images will become reality and these maps will become a type of standard products of photogrammetry.At last,we introduce our two current researches related to geometry and semantics respectively.One is stereo matching of aerial images based on deep learning and transfer learning;the other is precise crop classification from satellite spatio-temporal images based on 3D CNN.展开更多
Objective We aimed to compare the therapeutic effects of stereotactic aspiration and best medical management in patients who developed supratentorial hypertensive intracerebral haemorrhage(HICH)with a volume of haemor...Objective We aimed to compare the therapeutic effects of stereotactic aspiration and best medical management in patients who developed supratentorial hypertensive intracerebral haemorrhage(HICH)with a volume of haemorrhage between 20 and 40 mL.Methods The clinical data of 220 patients with supratentorial HICH with a volume between 20 and 40 mL were retrospectively analysed.Among them,142 received stereotactic aspiration surgery(stereotactic aspiration group)and 78 received best medical management(conservative group).All were followed up for 6 months.Multivariate logistic regression and Kaplan-Meier survival curves were used to compare the outcome between the two groups.results The rebleeding rate was lower in the group that had stereotactic aspiration when compared with the group with medical treatment(6[4.2%]vs 9[11.5%],χ2=4.364,p=0.037).After 6 months,although the mortality rate did not differ significantly between the two groups(8 cases[5.6%]vs 10 cases[12.8%],χ2=3.461,p=0.063),the rate of a favourable outcome was higher in the group who received stereotactic aspiration(χ2=15.870,p=0.000).Logistic regression identified that medical treatment(OR=1.64,p=0.000)was an independent risk factor for an unfavourable outcome.The Kaplan-Meier curves indicated that the median favourable outcome time in the stereotactic aspiration group was 59.5 days compared with that in the medically treated group(87.0 days).The log-rank test indicated that the prognosis at 6 months was better for those treated with stereotactic haematoma aspiration(χ2=29.866,p=0.000).However,the 6-month survival rate was similar between the two groups(χ2=3.253,p=0.068).Conclusions Stereotactic haematoma aspiration significantly improved the quality of life,although did not effectively reduce the rate of mortality.When selected appropriately,patients with HICH may benefit from this type of surgical intervention.展开更多
The currently available 3D printing still cannot simultaneously deal with the metal and nonmetal inks together due to their huge difference in the melting points and poor compatible printability between each other. He...The currently available 3D printing still cannot simultaneously deal with the metal and nonmetal inks together due to their huge difference in the melting points and poor compatible printability between each other. Here through introducing the low melting point alloy Bi35In48.6Sn16Zno.4 and silicone rubber as functional inks, we proposed a compatible hybrid 3D printing method for manufacturing the desired device, the supporting substrate and the allied package structure together. The principle of pneumatic-typed 3D printing of multiple inks was described and typical physical properties of the ink Bi35In48.6Sn16Zno.4 were measured. Several key factors dominating the printing quality such as the temperature of the printing head, the air pressure exerted upon the liquid metal ink in the syringe, the moving velocity and the height of the printing head etc. were clarified. A general way of directly printing out 3D structured electronic devices consisting of both metal and nonmetal materials was demonstrated. Such hybrid objects were patterned and formed up layer by layer with Bi35In48.6Sn16Zno.4 alloy and silicone rub- ber which would become solidified after standing for a period of time under room temperature. To illustrate the compatible printability of these printing inks, a three-layer tricolor LED stereo circuit with controlled lighting capability was further man- ufactured and evaluated. The present study opens an important hybrid 3D printing way for directly manufacturing functional and structural end devices in an easy and low cost way.展开更多
Plant phenotyping technologies play important roles in plant research and agriculture.Detailed phenotypes of individual plants can guide the optimization of shoot architecture for plant breeding and are useful to anal...Plant phenotyping technologies play important roles in plant research and agriculture.Detailed phenotypes of individual plants can guide the optimization of shoot architecture for plant breeding and are useful to analyze the morphological differences in response to environments for crop cultivation.Accordingly,high-throughput phenotyping technologies for individual plants grown in field conditions are urgently needed,and MVS-Pheno,a portable and low-cost phenotyping platform for individual plants,was developed.The platform is composed of four major components:a semiautomatic multiview stereo(MVS)image acquisition device,a data acquisition console,data processing and phenotype extraction software for maize shoots,and a data management system.The platform’s device is detachable and adjustable according to the size of the target shoot.Image sequences for each maize shoot can be captured within 60-120 seconds,yielding 3D point clouds of shoots are reconstructed using MVS-based commercial software,and the phenotypic traits at the organ and individual plant levels are then extracted by the software.The correlation coefficient(R^(2))between the extracted and manually measured plant height,leaf width,and leaf area values are 0.99,0.87,and 0.93,respectively.A data management system has also been developed to store and manage the acquired raw data,reconstructed point clouds,agronomic information,and resulting phenotypic traits.The platform offers an optional solution for high-throughput phenotyping of field-grown plants,which is especially useful for large populations or experiments across many different ecological regions.展开更多
A new motion model and estimation algorithm is proposed to compute the general rigid motion object＇s 6-DOF motion parameters and center of rotation based on stereo vision. The object＇s 6-DOF motion model is designed...A new motion model and estimation algorithm is proposed to compute the general rigid motion object＇s 6-DOF motion parameters and center of rotation based on stereo vision. The object＇s 6-DOF motion model is designed from the rigid object＇s motion character under the two defined reference frames. According to the rigid object＇s motion model and motion dynamics knowledge, the corresponding motion algorithm to compute the 6-DOF motion parameters is worked out. By the rigid object pure rotation motion model and space sphere geometry knowledge, the center of rotation may be calculated after eliminating the translation motion out of the 6-DOF motion. The motion equations are educed based on the motion model and the closed-form solutions are figured out. To heighten the motion estimation algorithm＇s robust, RANSAC algorithm is applied to delete the outliers. Simulation and real experiments are conducted and the experiment results are analyzed. The results prove the motion model＇s correction and algorithm＇s validity.展开更多
The borescopy inspection problem of aeroengine interior important partdamages such as firebox's burn and corruption, vane' s crack, bump, abrade and concave pit, is aimedat. A new system is developed to carry ...The borescopy inspection problem of aeroengine interior important partdamages such as firebox's burn and corruption, vane' s crack, bump, abrade and concave pit, is aimedat. A new system is developed to carry out 3D measurement and stereo reconstruction of engineinterior damage, in which the borescope of Japanese OLYMPUS Corporation is used as hardware. In thesystem, functions are implemented, such as image collection, camera calibration, imagepreprocessing, stereo matching, 3D measurement and stereo reconstruction. It can provide moredetailed inspection and more accurate estimation of engine interior damages. Finally, an example isused to verify the effectivity of the new method.展开更多
The goals of engineering and scientific missions for Chang＇E-2 lunar satellite require high detection sensitivity and large imaging dynamic range for the onboard CCD cameras. The TDI CCD image sensor was adopted for ...The goals of engineering and scientific missions for Chang＇E-2 lunar satellite require high detection sensitivity and large imaging dynamic range for the onboard CCD cameras. The TDI CCD image sensor was adopted for the two linear CCD stereo cameras for the first time in the lunar reconnaissance of the world. The design argumentation is described in this paper. The analysis shows that the imagers meet the mission requirements. The satellite was launched on 1 October 2010 at zero window. The cameras obtained images of 7 m resolution on the 100 km orbit for the first time on 24 October 2010, and operated once again on 27 October 2010 to take stereo images of the Sinus Iridum with the resolution better than 1.5 m. On the near-moon-arc of 15 kmxl00 km elliptical orbit, the images are very clear and rich of grey scales, indicating successful completion of the Chang＇E-2 engineering mission. At the present the cameras are acquiring the full lunar surface stereo images with 7 m resolution on the 100 km circular orbit to complete their scientific mission.展开更多
Growth process of iron whiskers and mechanism of CaO influence on precipitation morphology of metallic iron at the gas-solid interfaces was studied. Analytical reagents of Fe（NO3）3 and Ca（NO3）2 aqueous solution we...Growth process of iron whiskers and mechanism of CaO influence on precipitation morphology of metallic iron at the gas-solid interfaces was studied. Analytical reagents of Fe（NO3）3 and Ca（NO3）2 aqueous solution were used to prepare sheet film sample of Fe2 O3-CAO by thermal decomposition at high temperature. In-situ observation was con-ducted using a stereo optical microscope and a hot-stage. And reduction kinetics of samples was studied by thermo gravimetrie （TG） method. Some samples after reduction were analyzed by using the scanning electron microscope （SEM）, energy dispersive spectrometer （EDS） and fourier transform infrared （FT-IR） spectrometer. Results indi-cate that during the reduction of iron oxides with CO, metallic iron is mostly precipitated as whisker and the precipi- tation behavior mainly depends on reduction rate. Doping CaO can significantly increase the reduction rate and effec-tively change the precipitation morphology of metallic iron after the reduction. When CaO doping concentration is less than 4% （mass percent）, CaO can promote whisker formation of reduced iron; as it reaches 6% （mass per- cent）, CaO inhibits iron whiskers growth; as it is more than 8% （mass percent）, no whiskers could be observed. Therefore, controlling the quantity of Ca^2＋ is effective to control the formation and growth of iron whiskers during gaseous reduction and thus eliminating ore grain sticking caused by intertexture of iron whiskers.展开更多
基金National Natural Science Foundation of China(41471288).
文摘Deep learning has become popular and the mainstream technology in many researches related to learning,and has shown its impact on photogrammetry.According to the definition of photogrammetry,that is,a subject that researches shapes,locations,sizes,characteristics and inter-relationships of real objects from optical images,photogrammetry considers two aspects,geometry and semantics.From the two aspects,we review the history of deep learning and discuss its current applications on photogrammetry,and forecast the future development of photogrammetry.In geometry,the deep convolutional neural network(CNN)has been widely applied in stereo matching,SLAM and 3D reconstruction,and has made some effects but needs more improvement.In semantics,conventional methods that have to design empirical and handcrafted features have failed to extract the semantic information accurately and failed to produce types of“semantic thematic map”as 4D productions(DEM,DOM,DLG,DRG)of photogrammetry.This causes the semantic part of photogrammetry be ignored for a long time.The powerful generalization capacity,ability to fit any functions and stability under types of situations of deep leaning is making the automatic production of thematic maps possible.We review the achievements that have been obtained in road network extraction,building detection and crop classification,etc.,and forecast that producing high-accuracy semantic thematic maps directly from optical images will become reality and these maps will become a type of standard products of photogrammetry.At last,we introduce our two current researches related to geometry and semantics respectively.One is stereo matching of aerial images based on deep learning and transfer learning;the other is precise crop classification from satellite spatio-temporal images based on 3D CNN.
基金This study was supported by awards from the Science and Technology Department(Henan Province,China:182102310158)Education Department of Henan Province,China(201503135).
文摘Objective We aimed to compare the therapeutic effects of stereotactic aspiration and best medical management in patients who developed supratentorial hypertensive intracerebral haemorrhage(HICH)with a volume of haemorrhage between 20 and 40 mL.Methods The clinical data of 220 patients with supratentorial HICH with a volume between 20 and 40 mL were retrospectively analysed.Among them,142 received stereotactic aspiration surgery(stereotactic aspiration group)and 78 received best medical management(conservative group).All were followed up for 6 months.Multivariate logistic regression and Kaplan-Meier survival curves were used to compare the outcome between the two groups.results The rebleeding rate was lower in the group that had stereotactic aspiration when compared with the group with medical treatment(6[4.2%]vs 9[11.5%],χ2=4.364,p=0.037).After 6 months,although the mortality rate did not differ significantly between the two groups(8 cases[5.6%]vs 10 cases[12.8%],χ2=3.461,p=0.063),the rate of a favourable outcome was higher in the group who received stereotactic aspiration(χ2=15.870,p=0.000).Logistic regression identified that medical treatment(OR=1.64,p=0.000)was an independent risk factor for an unfavourable outcome.The Kaplan-Meier curves indicated that the median favourable outcome time in the stereotactic aspiration group was 59.5 days compared with that in the medically treated group(87.0 days).The log-rank test indicated that the prognosis at 6 months was better for those treated with stereotactic haematoma aspiration(χ2=29.866,p=0.000).However,the 6-month survival rate was similar between the two groups(χ2=3.253,p=0.068).Conclusions Stereotactic haematoma aspiration significantly improved the quality of life,although did not effectively reduce the rate of mortality.When selected appropriately,patients with HICH may benefit from this type of surgical intervention.
基金supported by the Research Funding of the Chinese Academy of Sciences(Grant No.KGZD-EW-T04-4)
文摘The currently available 3D printing still cannot simultaneously deal with the metal and nonmetal inks together due to their huge difference in the melting points and poor compatible printability between each other. Here through introducing the low melting point alloy Bi35In48.6Sn16Zno.4 and silicone rubber as functional inks, we proposed a compatible hybrid 3D printing method for manufacturing the desired device, the supporting substrate and the allied package structure together. The principle of pneumatic-typed 3D printing of multiple inks was described and typical physical properties of the ink Bi35In48.6Sn16Zno.4 were measured. Several key factors dominating the printing quality such as the temperature of the printing head, the air pressure exerted upon the liquid metal ink in the syringe, the moving velocity and the height of the printing head etc. were clarified. A general way of directly printing out 3D structured electronic devices consisting of both metal and nonmetal materials was demonstrated. Such hybrid objects were patterned and formed up layer by layer with Bi35In48.6Sn16Zno.4 alloy and silicone rub- ber which would become solidified after standing for a period of time under room temperature. To illustrate the compatible printability of these printing inks, a three-layer tricolor LED stereo circuit with controlled lighting capability was further man- ufactured and evaluated. The present study opens an important hybrid 3D printing way for directly manufacturing functional and structural end devices in an easy and low cost way.
基金This research was funded by the Construction of Collaborative Innovation Center of Beijing Academy of Agricultural and Forestry Sciences(KJCX201917)the National Natural Science Foundation of China(31871519 and 31601215)+1 种基金the Modern Agro-Industry Technology Research System of Maize(CARS-02-87)the Construction of Scientific Research and Innovation Platform in Beijing Academy of Agricultural and Forestry Sciences(Digital Plant).
文摘Plant phenotyping technologies play important roles in plant research and agriculture.Detailed phenotypes of individual plants can guide the optimization of shoot architecture for plant breeding and are useful to analyze the morphological differences in response to environments for crop cultivation.Accordingly,high-throughput phenotyping technologies for individual plants grown in field conditions are urgently needed,and MVS-Pheno,a portable and low-cost phenotyping platform for individual plants,was developed.The platform is composed of four major components:a semiautomatic multiview stereo(MVS)image acquisition device,a data acquisition console,data processing and phenotype extraction software for maize shoots,and a data management system.The platform’s device is detachable and adjustable according to the size of the target shoot.Image sequences for each maize shoot can be captured within 60-120 seconds,yielding 3D point clouds of shoots are reconstructed using MVS-based commercial software,and the phenotypic traits at the organ and individual plant levels are then extracted by the software.The correlation coefficient(R^(2))between the extracted and manually measured plant height,leaf width,and leaf area values are 0.99,0.87,and 0.93,respectively.A data management system has also been developed to store and manage the acquired raw data,reconstructed point clouds,agronomic information,and resulting phenotypic traits.The platform offers an optional solution for high-throughput phenotyping of field-grown plants,which is especially useful for large populations or experiments across many different ecological regions.
基金National Natural Science Foundation of China (No.50275040)
文摘A new motion model and estimation algorithm is proposed to compute the general rigid motion object＇s 6-DOF motion parameters and center of rotation based on stereo vision. The object＇s 6-DOF motion model is designed from the rigid object＇s motion character under the two defined reference frames. According to the rigid object＇s motion model and motion dynamics knowledge, the corresponding motion algorithm to compute the 6-DOF motion parameters is worked out. By the rigid object pure rotation motion model and space sphere geometry knowledge, the center of rotation may be calculated after eliminating the translation motion out of the 6-DOF motion. The motion equations are educed based on the motion model and the closed-form solutions are figured out. To heighten the motion estimation algorithm＇s robust, RANSAC algorithm is applied to delete the outliers. Simulation and real experiments are conducted and the experiment results are analyzed. The results prove the motion model＇s correction and algorithm＇s validity.
文摘The borescopy inspection problem of aeroengine interior important partdamages such as firebox's burn and corruption, vane' s crack, bump, abrade and concave pit, is aimedat. A new system is developed to carry out 3D measurement and stereo reconstruction of engineinterior damage, in which the borescope of Japanese OLYMPUS Corporation is used as hardware. In thesystem, functions are implemented, such as image collection, camera calibration, imagepreprocessing, stereo matching, 3D measurement and stereo reconstruction. It can provide moredetailed inspection and more accurate estimation of engine interior damages. Finally, an example isused to verify the effectivity of the new method.
文摘The goals of engineering and scientific missions for Chang＇E-2 lunar satellite require high detection sensitivity and large imaging dynamic range for the onboard CCD cameras. The TDI CCD image sensor was adopted for the two linear CCD stereo cameras for the first time in the lunar reconnaissance of the world. The design argumentation is described in this paper. The analysis shows that the imagers meet the mission requirements. The satellite was launched on 1 October 2010 at zero window. The cameras obtained images of 7 m resolution on the 100 km orbit for the first time on 24 October 2010, and operated once again on 27 October 2010 to take stereo images of the Sinus Iridum with the resolution better than 1.5 m. On the near-moon-arc of 15 kmxl00 km elliptical orbit, the images are very clear and rich of grey scales, indicating successful completion of the Chang＇E-2 engineering mission. At the present the cameras are acquiring the full lunar surface stereo images with 7 m resolution on the 100 km circular orbit to complete their scientific mission.
基金Item Sponsored by National Natural Science Foundation of China ( 50834007 )National Basic Research Program of China ( 2012CB720401 )
文摘Growth process of iron whiskers and mechanism of CaO influence on precipitation morphology of metallic iron at the gas-solid interfaces was studied. Analytical reagents of Fe（NO3）3 and Ca（NO3）2 aqueous solution were used to prepare sheet film sample of Fe2 O3-CAO by thermal decomposition at high temperature. In-situ observation was con-ducted using a stereo optical microscope and a hot-stage. And reduction kinetics of samples was studied by thermo gravimetrie （TG） method. Some samples after reduction were analyzed by using the scanning electron microscope （SEM）, energy dispersive spectrometer （EDS） and fourier transform infrared （FT-IR） spectrometer. Results indi-cate that during the reduction of iron oxides with CO, metallic iron is mostly precipitated as whisker and the precipi- tation behavior mainly depends on reduction rate. Doping CaO can significantly increase the reduction rate and effec-tively change the precipitation morphology of metallic iron after the reduction. When CaO doping concentration is less than 4% （mass percent）, CaO can promote whisker formation of reduced iron; as it reaches 6% （mass per- cent）, CaO inhibits iron whiskers growth; as it is more than 8% （mass percent）, no whiskers could be observed. Therefore, controlling the quantity of Ca^2＋ is effective to control the formation and growth of iron whiskers during gaseous reduction and thus eliminating ore grain sticking caused by intertexture of iron whiskers.
