The traditional single image dehazing algorithm is susceptible to the prior knowledge of hazy image and colour distortion.A new method of deep learning multi-scale convolution neural network based on HSI colour space ...The traditional single image dehazing algorithm is susceptible to the prior knowledge of hazy image and colour distortion.A new method of deep learning multi-scale convolution neural network based on HSI colour space for single image dehazing is proposed in this paper,which directly learns the mapping relationship between hazy image and corresponding clear image in colour,saturation and brightness by the designed structure of deep learning network to achieve haze removal.Firstly,the hazy image is transformed from RGB colour space to HSI colour space.Secondly,an end-to-end multi-scale full convolution neural network model is designed.The multi-scale extraction is realized by three different dehazing sub-networks:hue H,saturation S and intensity I,and the mapping relationship between hazy image and clear image is obtained by deep learning.Finally,the model was trained and tested with hazy data set.The experimental results show that this method can achieve good dehazing effect for both synthetic hazy images and real hazy images,and is superior to other contrast algorithms in subjective and objective evaluations.展开更多
An assumption that <em>all</em> the six flavour quarks are attributed to be the components of <em>a same, a</em> <em>common</em> isospin multiplets space named <strong>STS<...An assumption that <em>all</em> the six flavour quarks are attributed to be the components of <em>a same, a</em> <em>common</em> isospin multiplets space named <strong>STS</strong> is proposed. Base on <strong>Pauli Exclusion Principle</strong>, every quark is assigned to different flavour marks in STS. Every flavour quark possesses <em>its own colour spectral line array</em> specially appointed. The collection of colour spectral line arrays of the six flavour quarks constructs together the <strong>CSDF</strong>, Colour Spectrum Diagram of Flavour, further baryons and mesons could be constructed from <strong>CSDF</strong>. STS, Spin Topological Space is a math frame with infinite dimensional matrix representation for spin angular momentum. Flavours is an isospin angular momentum coupling phenomena of the three-colour-quarks.展开更多
Rewardless orchids attract pollinators by food, sexual, and brood-site mimicry, but other forms of sensory deception may also operate. Helmet orchids （Corybas, Nematoceras and related genera） are often assumed to be...Rewardless orchids attract pollinators by food, sexual, and brood-site mimicry, but other forms of sensory deception may also operate. Helmet orchids （Corybas, Nematoceras and related genera） are often assumed to be brood-site deceivers that mimic the colours and scents of mushrooms to fool female fungus gnats （Mycetophilidae） into attempting oviposition and polli- nating flowers. We sampled spectral reflectances and volatile odours of an endemic terrestrial New Zealand orchid Corybas cheesemanii, and co-occurring wild mushrooms. The orchid is scentless to humans and SPME GC-MS analyses did not detect any odours, but more sensitive methods may be required. The orchids reflected strongly across all visible wavelengths （300-700nm） with peaks in the UV （-320nm）, yellow-green （500-600 nm） and red regions （650-700 nm）, whereas mushrooms and surrounding leaf litter reflected predominantly red and no UV. Rather than mimicking mushrooms, these orchids may attract pollinators by exploiting insects＇ strong sensory bias for UV. Modelling spectral reflectances into a categorical fly vision model and a generic tetrachromat vision model provided very different results, but neither suggest any mimicry of mushrooms. However, these models require further assessment and data on fly spectral sensitivity to red wavelengths is lacking - a problem given the predominance of red, fly-pollinated flowers worldwide展开更多
基金National Natural Science Foundation of China(No.61963023)MOE(Ministry of Education in China)Project of Humanities and Social Sciences(No.19YJC760012)。
文摘The traditional single image dehazing algorithm is susceptible to the prior knowledge of hazy image and colour distortion.A new method of deep learning multi-scale convolution neural network based on HSI colour space for single image dehazing is proposed in this paper,which directly learns the mapping relationship between hazy image and corresponding clear image in colour,saturation and brightness by the designed structure of deep learning network to achieve haze removal.Firstly,the hazy image is transformed from RGB colour space to HSI colour space.Secondly,an end-to-end multi-scale full convolution neural network model is designed.The multi-scale extraction is realized by three different dehazing sub-networks:hue H,saturation S and intensity I,and the mapping relationship between hazy image and clear image is obtained by deep learning.Finally,the model was trained and tested with hazy data set.The experimental results show that this method can achieve good dehazing effect for both synthetic hazy images and real hazy images,and is superior to other contrast algorithms in subjective and objective evaluations.
文摘An assumption that <em>all</em> the six flavour quarks are attributed to be the components of <em>a same, a</em> <em>common</em> isospin multiplets space named <strong>STS</strong> is proposed. Base on <strong>Pauli Exclusion Principle</strong>, every quark is assigned to different flavour marks in STS. Every flavour quark possesses <em>its own colour spectral line array</em> specially appointed. The collection of colour spectral line arrays of the six flavour quarks constructs together the <strong>CSDF</strong>, Colour Spectrum Diagram of Flavour, further baryons and mesons could be constructed from <strong>CSDF</strong>. STS, Spin Topological Space is a math frame with infinite dimensional matrix representation for spin angular momentum. Flavours is an isospin angular momentum coupling phenomena of the three-colour-quarks.
文摘Rewardless orchids attract pollinators by food, sexual, and brood-site mimicry, but other forms of sensory deception may also operate. Helmet orchids （Corybas, Nematoceras and related genera） are often assumed to be brood-site deceivers that mimic the colours and scents of mushrooms to fool female fungus gnats （Mycetophilidae） into attempting oviposition and polli- nating flowers. We sampled spectral reflectances and volatile odours of an endemic terrestrial New Zealand orchid Corybas cheesemanii, and co-occurring wild mushrooms. The orchid is scentless to humans and SPME GC-MS analyses did not detect any odours, but more sensitive methods may be required. The orchids reflected strongly across all visible wavelengths （300-700nm） with peaks in the UV （-320nm）, yellow-green （500-600 nm） and red regions （650-700 nm）, whereas mushrooms and surrounding leaf litter reflected predominantly red and no UV. Rather than mimicking mushrooms, these orchids may attract pollinators by exploiting insects＇ strong sensory bias for UV. Modelling spectral reflectances into a categorical fly vision model and a generic tetrachromat vision model provided very different results, but neither suggest any mimicry of mushrooms. However, these models require further assessment and data on fly spectral sensitivity to red wavelengths is lacking - a problem given the predominance of red, fly-pollinated flowers worldwide
