Fusarium head blight (FHB) caused by Fusarium graminearum is a devastating disease that results in extensive yield losses to wheat and barley. A green fluorescent protein (GFP) expressing plasmid pRP22-GFP was constru...Fusarium head blight (FHB) caused by Fusarium graminearum is a devastating disease that results in extensive yield losses to wheat and barley. A green fluorescent protein (GFP) expressing plasmid pRP22-GFP was constructed for monitoring the colonization of two biocontrol agents, Brevibacillus brevis ZJY-1 and Bacillus subtilis ZJY-116, on the spikes of barley and their effect on suppression of FHB. Survival and colonization of the Brevibacillus brevis ZJY-1 and Bacillus subtilis ZJY-116 strains on spikes of barley were observed by tracking the bacterial transformants with GFP expression. Our field study revealed that plasmid pRP22-GFP was stably maintained in the bacterial strains without selective pressure. The retrieved GFP-tagged strains showed that the bacterial population fluctuation accorded with that of the rain events. Furthermore, both biocontrol strains gave significant protection against FHB on spikes of barley in fields. The greater suppression of barley FHB disease was resulted from the treat-ment of barley spikes with biocontrol agents before inoculation with F. graminearum.展开更多
Epileptic seizures are known for their unpredictable nature.However,recent research provides that the transition to seizure event is not random but the result of evidence accumulations.Therefore,a reliable method capa...Epileptic seizures are known for their unpredictable nature.However,recent research provides that the transition to seizure event is not random but the result of evidence accumulations.Therefore,a reliable method capable to detect these indications can predict seizures and improve the life quality of epileptic patients.Seizures periods are generally characterized by epileptiform discharges with different changes including spike rate variation according to the shapes,spikes,and the amplitude.In this study,spike rate is used as the indicator to anticipate seizures in electroencephalogram(EEG) signal.Spikes detection step is used in EEG signal during interictal,preictal,and ictal periods followed by a mean filter to smooth the spike number.The maximum spike rate in interictal periods is used as an indicator to predict seizures.When the spike number in the preictal period exceeds the threshold,an alarm is triggered.Using the CHB-MIT database,the proposed approach has ensured92% accuracy in seizure prediction for all patients.展开更多
This paper investigates the finite-thickness effect of two superimposed fluids on bubbles and spikes in Richtmyer–Meshkov instability(RMI) for arbitrary Atwood numbers by using the method of the small parameter expan...This paper investigates the finite-thickness effect of two superimposed fluids on bubbles and spikes in Richtmyer–Meshkov instability(RMI) for arbitrary Atwood numbers by using the method of the small parameter expansion up to the second order. When the thickness of the two fluids tends to be infinity, our results can reproduce the classical results where RMI happens at the interface separating two semi-infinity-thickness fluids of different densities. It is found that the thickness has a large influence on the amplitude evolution of bubbles and spikes compared with those in classical RMI. Based on the thickness relationship of the two fluids, the thickness effect on bubbles and spikes for four cases is discussed. The thickness encourages(or reduces)the growth of bubbles or spikes, depending on not only Atwood number, but also the relationship of the thickness ratio of the heavy and light fluids, which is explicitly determined in this paper.展开更多
Data processing for seismic network is very complex and fussy,because a lot of data is recorded in seismicnetwork every day,which make it impossible to process these data all by manual work.Therefore,seismic datashoul...Data processing for seismic network is very complex and fussy,because a lot of data is recorded in seismicnetwork every day,which make it impossible to process these data all by manual work.Therefore,seismic datashould be processed automatically to produce a initial results about events detection and location.Afterwards,these results are reviewed and modified by analyst.In automatic processing data quality checking is important.There are three main problem data that exist in real seismic records,which include:spike,repeated data and展开更多
BACKGROUND Epilepsy is a syndrome characterized by transient,rigid,paroxysmal,and repetitive central nervous system dysfunction.Prevention,control,and improvement of cognitive and behavioral dysfunction are of great s...BACKGROUND Epilepsy is a syndrome characterized by transient,rigid,paroxysmal,and repetitive central nervous system dysfunction.Prevention,control,and improvement of cognitive and behavioral dysfunction are of great significance for improving the patients’intellectual development and quality of life.Electroencephalograms(EEG)can predict an accelerated decline in cognitive function.AIM To determine the clinical and EEG characteristics and treatment results of benign epilepsy in spiking children.METHODS A total of 106 cases of benign epilepsy in children with myocardial spines treated at our hospital from January 2017 to January 2020 were selected.Differences in clinical data and EGG characteristics between treatment-effective/-ineffective patients were analyzed,and children’s intellectual development before and after treatment evaluated using the Gesell Development Diagnostic Scale.RESULTS EEG showed that the discharge proportion in the awake and sleep periods was 66.04%,and the peak/peak discharge was mainly single-sided,accounting for 81.13%,while the discharge generalization accounted for 31.13%.There was no significant difference in any of these variables between sexes and ages(P>0.05).The proportion of patients with early onset(<5 years old)and seizure frequency>3 times/half a year was 40.00%and 60.00%,respectively;the incidence rate and seizure frequency in the younger age group(<5 years old)were significantly higher than those in the treatment-effective group(P<0.05),while the discharge index was significantly lower than that in the treatment-effective group(P<0.05).The discharge index was negatively correlated with fine motor skill and language development(r=-0.274 and-0.247,respectively;P<0.05),but not with the rest(P>0.05).Logistic regression analysis showed that low age onset(<5 years old)and seizure frequency were the factors affecting ineffective-treatment of benign epilepsy in children(odds ratio=11.304 and 5.784,respectively;P<0.05).The discharge index of the responsive group after treatment was significantly lower than that of the unresponsive group(P<0.05).However,there was no significant difference between groups after treatment in gross and fine motor skills,adaptability,language,and personal social development(P>0.05).CONCLUSION The EEG of children with benign epilepsy due to spinal wave in central time zone has characteristic changes,and the therapeutic effect is influenced by age of onset and attack frequency.展开更多
This paper presents a spikes removing methodology for ultrasonic rangefinders with an application to a quadrotor unmanned aerial vehicle. Ultrasonic sensors suffer from spikes in distance measurements due to specular ...This paper presents a spikes removing methodology for ultrasonic rangefinders with an application to a quadrotor unmanned aerial vehicle. Ultrasonic sensors suffer from spikes in distance measurements due to specular reflectance and acoustic noise. Removing these spikes is necessary for improving the hovering performance of the quadrotor. The spikes removing algorithm is based on the discrete wavelet transform. The algorithm is implemented in simulation to study the effect of the altitude measurement spikes on the control performance of the quadrotor with and without the algorithm. The algorithm is also implemented digitally on ultrasonic measurements from a real flight. Results show that the method is capable of rejecting the spikes in the measurements efficiently leaving the altitude control signal unaffected.展开更多
The characteristics of the millisecond spikes with short durations and weak flux densities which were observed with high time resolution (1 ms) during a great type Ⅳ solar radio burst of 1990 July 30 are introduced. ...The characteristics of the millisecond spikes with short durations and weak flux densities which were observed with high time resolution (1 ms) during a great type Ⅳ solar radio burst of 1990 July 30 are introduced. The time profiles of the spikes are statistically analyzed and the parameters of the spike source are also estimated.展开更多
Tangrams奖前身为亚洲营销实效和策略奖,如今为亚洲顶尖创意奖(Spikes Asia Awards)下设的一个主要奖项,包括实效类、媒介策略类、数字策略类、大数据和分析以及电子商务类几个类别。作为亚太区最大规模的策略和实效类奖项,Tangrams表...Tangrams奖前身为亚洲营销实效和策略奖,如今为亚洲顶尖创意奖(Spikes Asia Awards)下设的一个主要奖项,包括实效类、媒介策略类、数字策略类、大数据和分析以及电子商务类几个类别。作为亚太区最大规模的策略和实效类奖项,Tangrams表彰的是活动作品中所展现的先进营销实效思维,以及最出色的有效品牌构建和稳健的商业转化结果,在创意卓越之外,又为行业提供了营销策略和实效方面的基准。展开更多
When a mass spreads in a turbulent flow, areas with obviously high concentration of the mass compared with surrounding areas are formed by organized structures of turbulence. In this study, we extract the high concent...When a mass spreads in a turbulent flow, areas with obviously high concentration of the mass compared with surrounding areas are formed by organized structures of turbulence. In this study, we extract the high concentration areas and investigate their diffusion process. For this purpose, a combination of Planar Laser Induced Fluorescence (PLIF) and Particle Image Velocimetry (PIV) techniques was employed to obtain simultaneously the two fields of the concentration of injected dye and of the velocity in a water turbulent channel flow. With focusing on a quasi-homogeneous turbulence in the channel central region, a series of PLIF and PIV images were acquired at several different downstream positions. We applied a conditional sampling technique to the PLIF images to extract the high concentration areas, or spikes, and calculated the conditional-averaged statistics of the extracted areas such as length scale, mean concentration, and turbulent diffusion coefficient. We found that the averaged length scale was constant with downstream distance from the diffusion source and was smaller than integral scale of the turbulent eddies. The spanwise distribution of the mean concentration was basically Gaussian, and the spanwise width of the spikes increased linearly with downstream distance from the diffusion source. Moreover, the turbulent diffusion coefficient was found to increase in proportion to the spanwise distance from the source. These results reveal aspects different from those of regular mass diffusion and let us conclude that the diffusion process of the spikes differs from that of regular mass diffusion.展开更多
In recent years, blind source separation (BSS) by independent component analysis (ICA) has been drawing much attention because of its potential applications in signal processing such as in speech recognition syste...In recent years, blind source separation (BSS) by independent component analysis (ICA) has been drawing much attention because of its potential applications in signal processing such as in speech recognition systems, telecommunication and medical signal processing. In this paper, two algorithms of independent component analysis (fixed-point IC,4 and natural gradient-flexible ICA) are adopted to extract human epileptic feature spikes from interferential signals. Experiment results show that epileptic spikes can be extracted from noise successfully. The kurtosis of the epileptic component signal separated is much better than that of other noisy signals. It shows that ICA is an effective tool to extract epileptic spikes from patients' electroencephalogram EEG and shows promising application to assist physicians to diagnose epilepsy and estimate the epileptogenic region in clinic.展开更多
To evaluate the pollutant dispersion in background turbulent flows, most researches focus on statistical variation of concentrations or its fluctuations. However, those time-averaged quantities may be insufficient for...To evaluate the pollutant dispersion in background turbulent flows, most researches focus on statistical variation of concentrations or its fluctuations. However, those time-averaged quantities may be insufficient for risk assessment, because there emerge many high-intensity pollutant areas in the instantaneous concentration field. In this study, we tried to estimate the frequency of appearance of the high concentration areas in a turbulent flow based on the Probability Density Function (PDF) of concentration. The high concentration area was recognized by two conditions based on the concentration and the concentration gradient values. We considered that the estimation equation for the frequency of appearance of the recognized areas consisted of two terms based on each condition. In order to represent the two terms with physical quantities of velocity and concentration fields, simultaneous PIV (Particle Image Velocimetry) and PLIF (Planar Laser-Induced Fluorescence) measurement and PLIF time-serial measurement were performed in a quasi-homogeneous turbulent flow. According to the experimental results, one of the terms, related to the condition of the concentration, was found to be represented by the concentration PDF, while the other term, by the streamwise mean velocity and the integral length scale of the turbulent flow. Based on the results, we developed an estimation equation including the concentration PDF and the flow features of mean velocity and integral scale of turbulence. In the area where the concentration PDF was a Gaussian one, the difference between the frequencies of appearance estimated by the equation and calculated from the experimental data was within 25%, which showed good accuracy of our proposed estimation equation. Therefore, our proposed equation is feasible for estimating the frequency of appearance of high concentration areas in a limited area in turbulent mass diffusion.展开更多
“很多人调侃自己是‘广告狗’,但是我不觉得,我觉得做广告非常嗨。”完成2019戛纳幼狮中国区选拔赛暨2019 Young Spikes青年赛中国区选拔赛(以下简称“选拔赛”)评审工作后,其中一位评委在“伯乐大讲堂”对所有的选手说出了这番话,希...“很多人调侃自己是‘广告狗’,但是我不觉得,我觉得做广告非常嗨。”完成2019戛纳幼狮中国区选拔赛暨2019 Young Spikes青年赛中国区选拔赛(以下简称“选拔赛”)评审工作后,其中一位评委在“伯乐大讲堂”对所有的选手说出了这番话,希望用自己的经历激励大家在今后的广告事业中不忘初心、保持热爱,这也是选拔赛评审团“黄金天团”、选拔赛独家赞助商REDS,以及选拔赛主办方《现代广告》对所有选手的一致期许。展开更多
Spike development directly affects the yield and quality of rice. We describe an algorithm for automatically identifying multiple developmental stages of rice spikes(AI-MDSRS) that transforms the automatic identificat...Spike development directly affects the yield and quality of rice. We describe an algorithm for automatically identifying multiple developmental stages of rice spikes(AI-MDSRS) that transforms the automatic identification of multiple developmental stages of rice spikes into the detection of rice spikes of diverse maturity levels. The scales vary greatly in different growth and development stages because rice spikes are dense and small, posing challenges for their effective and accurate detection. We describe a rice spike detection model based on an improved faster regions with convolutional neural network(Faster R-CNN).The model incorporates the following optimization strategies: first, Inception_Res Net-v2 replaces VGG16 as a feature extraction network;second, a feature pyramid network(FPN) replaces single-scale feature maps to fuse with region proposal network(RPN);third, region of interest(Ro I) alignment replaces Ro I pooling, and distance-intersection over union(DIo U) is used as a standard for non-maximum suppression(NMS). The performance of the proposed model was compared with that of the original Faster R-CNN and YOLOv4 models. The mean average precision(m AP) of the rice spike detection model was92.47%, a substantial improvement on the original Faster R-CNN model(with 40.96% m AP) and 3.4%higher than that of the YOLOv4 model, experimentally indicating that the model is more accurate and reliable. The identification results of the model for the heading–flowering, milky maturity, and full maturity stages were within two days of the results of manual observation, fully meeting the needs of agricultural activities.展开更多
基金Project supported by the National Natural Science Foundation of China (No. 30230250)Science and Technology Committee of Zhejiang Province (No. 2003C22029), China
文摘Fusarium head blight (FHB) caused by Fusarium graminearum is a devastating disease that results in extensive yield losses to wheat and barley. A green fluorescent protein (GFP) expressing plasmid pRP22-GFP was constructed for monitoring the colonization of two biocontrol agents, Brevibacillus brevis ZJY-1 and Bacillus subtilis ZJY-116, on the spikes of barley and their effect on suppression of FHB. Survival and colonization of the Brevibacillus brevis ZJY-1 and Bacillus subtilis ZJY-116 strains on spikes of barley were observed by tracking the bacterial transformants with GFP expression. Our field study revealed that plasmid pRP22-GFP was stably maintained in the bacterial strains without selective pressure. The retrieved GFP-tagged strains showed that the bacterial population fluctuation accorded with that of the rain events. Furthermore, both biocontrol strains gave significant protection against FHB on spikes of barley in fields. The greater suppression of barley FHB disease was resulted from the treat-ment of barley spikes with biocontrol agents before inoculation with F. graminearum.
文摘Epileptic seizures are known for their unpredictable nature.However,recent research provides that the transition to seizure event is not random but the result of evidence accumulations.Therefore,a reliable method capable to detect these indications can predict seizures and improve the life quality of epileptic patients.Seizures periods are generally characterized by epileptiform discharges with different changes including spike rate variation according to the shapes,spikes,and the amplitude.In this study,spike rate is used as the indicator to anticipate seizures in electroencephalogram(EEG) signal.Spikes detection step is used in EEG signal during interictal,preictal,and ictal periods followed by a mean filter to smooth the spike number.The maximum spike rate in interictal periods is used as an indicator to predict seizures.When the spike number in the preictal period exceeds the threshold,an alarm is triggered.Using the CHB-MIT database,the proposed approach has ensured92% accuracy in seizure prediction for all patients.
基金supported by National Natural Science Foundation of China (Nos. U1530261,91852203,and 11472278)the Innovation Fund of Fundamental Technology Institute of All Value In Creation (No. JCY2015A005)+2 种基金the Natural Science Foundation of Sichuan Province (Nos. 18ZA0260,and 2018JY0454)the Natural Science Foundation of Mianyang Normal University (Nos. HX2017007,MYSY2017JC06 and MYSY2018T004)the National High-Tech Inertial Confinement Fusion Committee
文摘This paper investigates the finite-thickness effect of two superimposed fluids on bubbles and spikes in Richtmyer–Meshkov instability(RMI) for arbitrary Atwood numbers by using the method of the small parameter expansion up to the second order. When the thickness of the two fluids tends to be infinity, our results can reproduce the classical results where RMI happens at the interface separating two semi-infinity-thickness fluids of different densities. It is found that the thickness has a large influence on the amplitude evolution of bubbles and spikes compared with those in classical RMI. Based on the thickness relationship of the two fluids, the thickness effect on bubbles and spikes for four cases is discussed. The thickness encourages(or reduces)the growth of bubbles or spikes, depending on not only Atwood number, but also the relationship of the thickness ratio of the heavy and light fluids, which is explicitly determined in this paper.
基金National Natural Science Foundation of China (60172026).
文摘Data processing for seismic network is very complex and fussy,because a lot of data is recorded in seismicnetwork every day,which make it impossible to process these data all by manual work.Therefore,seismic datashould be processed automatically to produce a initial results about events detection and location.Afterwards,these results are reviewed and modified by analyst.In automatic processing data quality checking is important.There are three main problem data that exist in real seismic records,which include:spike,repeated data and
文摘BACKGROUND Epilepsy is a syndrome characterized by transient,rigid,paroxysmal,and repetitive central nervous system dysfunction.Prevention,control,and improvement of cognitive and behavioral dysfunction are of great significance for improving the patients’intellectual development and quality of life.Electroencephalograms(EEG)can predict an accelerated decline in cognitive function.AIM To determine the clinical and EEG characteristics and treatment results of benign epilepsy in spiking children.METHODS A total of 106 cases of benign epilepsy in children with myocardial spines treated at our hospital from January 2017 to January 2020 were selected.Differences in clinical data and EGG characteristics between treatment-effective/-ineffective patients were analyzed,and children’s intellectual development before and after treatment evaluated using the Gesell Development Diagnostic Scale.RESULTS EEG showed that the discharge proportion in the awake and sleep periods was 66.04%,and the peak/peak discharge was mainly single-sided,accounting for 81.13%,while the discharge generalization accounted for 31.13%.There was no significant difference in any of these variables between sexes and ages(P>0.05).The proportion of patients with early onset(<5 years old)and seizure frequency>3 times/half a year was 40.00%and 60.00%,respectively;the incidence rate and seizure frequency in the younger age group(<5 years old)were significantly higher than those in the treatment-effective group(P<0.05),while the discharge index was significantly lower than that in the treatment-effective group(P<0.05).The discharge index was negatively correlated with fine motor skill and language development(r=-0.274 and-0.247,respectively;P<0.05),but not with the rest(P>0.05).Logistic regression analysis showed that low age onset(<5 years old)and seizure frequency were the factors affecting ineffective-treatment of benign epilepsy in children(odds ratio=11.304 and 5.784,respectively;P<0.05).The discharge index of the responsive group after treatment was significantly lower than that of the unresponsive group(P<0.05).However,there was no significant difference between groups after treatment in gross and fine motor skills,adaptability,language,and personal social development(P>0.05).CONCLUSION The EEG of children with benign epilepsy due to spinal wave in central time zone has characteristic changes,and the therapeutic effect is influenced by age of onset and attack frequency.
文摘This paper presents a spikes removing methodology for ultrasonic rangefinders with an application to a quadrotor unmanned aerial vehicle. Ultrasonic sensors suffer from spikes in distance measurements due to specular reflectance and acoustic noise. Removing these spikes is necessary for improving the hovering performance of the quadrotor. The spikes removing algorithm is based on the discrete wavelet transform. The algorithm is implemented in simulation to study the effect of the altitude measurement spikes on the control performance of the quadrotor with and without the algorithm. The algorithm is also implemented digitally on ultrasonic measurements from a real flight. Results show that the method is capable of rejecting the spikes in the measurements efficiently leaving the altitude control signal unaffected.
文摘The characteristics of the millisecond spikes with short durations and weak flux densities which were observed with high time resolution (1 ms) during a great type Ⅳ solar radio burst of 1990 July 30 are introduced. The time profiles of the spikes are statistically analyzed and the parameters of the spike source are also estimated.
文摘Tangrams奖前身为亚洲营销实效和策略奖,如今为亚洲顶尖创意奖(Spikes Asia Awards)下设的一个主要奖项,包括实效类、媒介策略类、数字策略类、大数据和分析以及电子商务类几个类别。作为亚太区最大规模的策略和实效类奖项,Tangrams表彰的是活动作品中所展现的先进营销实效思维,以及最出色的有效品牌构建和稳健的商业转化结果,在创意卓越之外,又为行业提供了营销策略和实效方面的基准。
文摘When a mass spreads in a turbulent flow, areas with obviously high concentration of the mass compared with surrounding areas are formed by organized structures of turbulence. In this study, we extract the high concentration areas and investigate their diffusion process. For this purpose, a combination of Planar Laser Induced Fluorescence (PLIF) and Particle Image Velocimetry (PIV) techniques was employed to obtain simultaneously the two fields of the concentration of injected dye and of the velocity in a water turbulent channel flow. With focusing on a quasi-homogeneous turbulence in the channel central region, a series of PLIF and PIV images were acquired at several different downstream positions. We applied a conditional sampling technique to the PLIF images to extract the high concentration areas, or spikes, and calculated the conditional-averaged statistics of the extracted areas such as length scale, mean concentration, and turbulent diffusion coefficient. We found that the averaged length scale was constant with downstream distance from the diffusion source and was smaller than integral scale of the turbulent eddies. The spanwise distribution of the mean concentration was basically Gaussian, and the spanwise width of the spikes increased linearly with downstream distance from the diffusion source. Moreover, the turbulent diffusion coefficient was found to increase in proportion to the spanwise distance from the source. These results reveal aspects different from those of regular mass diffusion and let us conclude that the diffusion process of the spikes differs from that of regular mass diffusion.
基金973 Project (No. 2003CB71606) and National Natural Science Foundation of China (No.30400105, 90208003)
文摘In recent years, blind source separation (BSS) by independent component analysis (ICA) has been drawing much attention because of its potential applications in signal processing such as in speech recognition systems, telecommunication and medical signal processing. In this paper, two algorithms of independent component analysis (fixed-point IC,4 and natural gradient-flexible ICA) are adopted to extract human epileptic feature spikes from interferential signals. Experiment results show that epileptic spikes can be extracted from noise successfully. The kurtosis of the epileptic component signal separated is much better than that of other noisy signals. It shows that ICA is an effective tool to extract epileptic spikes from patients' electroencephalogram EEG and shows promising application to assist physicians to diagnose epilepsy and estimate the epileptogenic region in clinic.
文摘To evaluate the pollutant dispersion in background turbulent flows, most researches focus on statistical variation of concentrations or its fluctuations. However, those time-averaged quantities may be insufficient for risk assessment, because there emerge many high-intensity pollutant areas in the instantaneous concentration field. In this study, we tried to estimate the frequency of appearance of the high concentration areas in a turbulent flow based on the Probability Density Function (PDF) of concentration. The high concentration area was recognized by two conditions based on the concentration and the concentration gradient values. We considered that the estimation equation for the frequency of appearance of the recognized areas consisted of two terms based on each condition. In order to represent the two terms with physical quantities of velocity and concentration fields, simultaneous PIV (Particle Image Velocimetry) and PLIF (Planar Laser-Induced Fluorescence) measurement and PLIF time-serial measurement were performed in a quasi-homogeneous turbulent flow. According to the experimental results, one of the terms, related to the condition of the concentration, was found to be represented by the concentration PDF, while the other term, by the streamwise mean velocity and the integral length scale of the turbulent flow. Based on the results, we developed an estimation equation including the concentration PDF and the flow features of mean velocity and integral scale of turbulence. In the area where the concentration PDF was a Gaussian one, the difference between the frequencies of appearance estimated by the equation and calculated from the experimental data was within 25%, which showed good accuracy of our proposed estimation equation. Therefore, our proposed equation is feasible for estimating the frequency of appearance of high concentration areas in a limited area in turbulent mass diffusion.
文摘“很多人调侃自己是‘广告狗’,但是我不觉得,我觉得做广告非常嗨。”完成2019戛纳幼狮中国区选拔赛暨2019 Young Spikes青年赛中国区选拔赛(以下简称“选拔赛”)评审工作后,其中一位评委在“伯乐大讲堂”对所有的选手说出了这番话,希望用自己的经历激励大家在今后的广告事业中不忘初心、保持热爱,这也是选拔赛评审团“黄金天团”、选拔赛独家赞助商REDS,以及选拔赛主办方《现代广告》对所有选手的一致期许。
基金supported by the Key-Area Research and Development Program of Guangdong Province (2019B020214005)Agricultural Research Project and Agricultural Technology Promotion Project of Guangdong (2021KJ383)。
文摘Spike development directly affects the yield and quality of rice. We describe an algorithm for automatically identifying multiple developmental stages of rice spikes(AI-MDSRS) that transforms the automatic identification of multiple developmental stages of rice spikes into the detection of rice spikes of diverse maturity levels. The scales vary greatly in different growth and development stages because rice spikes are dense and small, posing challenges for their effective and accurate detection. We describe a rice spike detection model based on an improved faster regions with convolutional neural network(Faster R-CNN).The model incorporates the following optimization strategies: first, Inception_Res Net-v2 replaces VGG16 as a feature extraction network;second, a feature pyramid network(FPN) replaces single-scale feature maps to fuse with region proposal network(RPN);third, region of interest(Ro I) alignment replaces Ro I pooling, and distance-intersection over union(DIo U) is used as a standard for non-maximum suppression(NMS). The performance of the proposed model was compared with that of the original Faster R-CNN and YOLOv4 models. The mean average precision(m AP) of the rice spike detection model was92.47%, a substantial improvement on the original Faster R-CNN model(with 40.96% m AP) and 3.4%higher than that of the YOLOv4 model, experimentally indicating that the model is more accurate and reliable. The identification results of the model for the heading–flowering, milky maturity, and full maturity stages were within two days of the results of manual observation, fully meeting the needs of agricultural activities.