General object recognition in complex backgrounds is still challenging. On one hand, the various backgrounds, where object may appear at different locations, make it difficult to find the object of interest. On the other hand, with the numbers of locations, types and variations in each type (e.g., rotation) increasing, conventional model-based approaches start to break down. The Where-What Networks (WWNs) were a biologically inspired framework for recognizing learned objects (appearances) from complex backgrounds. However, they do not have an adaptive receptive field for an object of a curved contour. Leaked-in background pixels will cause problems when different objects look similar. This work introduces a new biologically inspired mechanism -synapse maintenance and uses both supervised (motor-supervised for class response) and unsupervised learning (synapse maintenance) to realize objects recognition. Synapse maintenance is meant to automatically decide which synapse should be active firing of the postsynaptic neuron. With the synapse maintenance, the network has achieved a significant improvement in the network performance.
Bio‐tillage has recently been proposed as a measure to alleviate soil compaction through biopores created by cover crop roots. The objective of this study was to determine the effect of different cover crops on soil physical properties and the succeeding maize (Zea mays L.) growth in compacted soil. Four treatments, including no cover crop as a control (Con), alfalfa (Medicago sativa L.), oilseed rape (Brassica napus L.), and radish and hairy vetch mixture (Raphanus sativus L. and Vicia villosa Roth), were carried out under both compacted and noncompacted soil conditions. Soil physical properties, such as the volumetric soil water content (SWC), bulk density, saturated hydraulic conductivity (Ks) and air permeability at water potential of −60 hPa (Ka60), and maize root characteristics and yield were measured. The cover crops did not affect the soil bulk density but significantly decreased the SWC in both the compacted and noncompacted soils relative to the Con treatment. The alfalfa treatment presented significantly higher Ks in the noncompacted soil and Ka60 in both the compacted and noncompacted soils than the Con treatment in the soil layer depth of 20–50 cm. The three cover crop treatments improved the maize root biomass density (173.2% for 2018 and 35.6% for 2019) and root length density (50.9% for 2018 and 51.8% for 2019) relative to the Con treatment in the soil layer depth of 10–70 cm in 2018 and soil layer depth of 10–50 cm in 2019 in the compacted soil rather than in the noncompacted soil. Compared with the Con treatment, the radish mixed with hairy vetch treatment in 2018 and the oilseed rape treatment in 2019 significantly enhanced the maize yield in the compacted soil. Our results suggest that alfalfa is the best crop for improving air permeability; however, the oilseed rape and mixture of radish and hairy vetch lead to better maize growth in the compacted soil. Bio‐tillage using cover crops is effective in alleviating soil compaction.
Abstract-While a physical environment interacts with a human individual through the brain's sensors and effectors, internal representations inside the skull-closed brain autonomously emerge and adapt throughout the lifetime. By "skull-closed", we mean that the brain inside the skull is off limit to all teachers in the external physical environment, except the brain's sensory ends and motor ends. We present the Where-What Network 6 (WWN-6), which has realized our goal of fully autonomous development inside a closed network "skull". This means that the human programmer is not allowed to handcraft the internal representation for any fixed extra-body concepts. For example, the meanings of specific values of location or type concept are not known during the programming time. Such meanings emerge through associations imbedded in 'postnatal" experience. This capability is especially challenging when one considers the fact that most elements in the sensory ends are irrelevant to the signals at the effector ends (e.g., many background pixels). How does each vector value in the effector find its corresponding pattern in the correct patch of the sensory image? We outline this autonomous learning theory for the brain and present how the developmental program (DP) of WWN-6 enables the network to perform for attending and recognizing objects in complex backgrounds using natural video. The inputs to the agent (i.e., the network) are not artificially synthesized images as WWNs used before, but drawn from continuous video taken from natural settings where, in general, everything is moving.
Soil aeration is critical for crop growth, which is generally assessed by air-filled porosity (AFP). In non-rigid soils with high shrinkage and swelling, the AFP is not only related to the total porosity and soil moisture, but also to the soil shrinkage behaviour. However, the relationship between AFP and soil shrinkage has not been clarified. The objectives of this study were to (1) establish a mathematical equation to describe the behaviours of AFP associated with soil shrinkage (AFP sh ); and (2) to evaluate the influence of tillage practice on the AFP sh . Undisturbed soil core samples were collected from the 0-10 cm to 10-20 cm layers in a Vertisol under four tillage treatments (No tillage, NT; Rotary tillage, RT; Subsoiling, SS; Deep ploughing, DP) to measure the changes in the volumes of soil core, soil water and air at different soil matric potentials. Our results showed that an equation of AFP sh as a function of moisture ratio (ϑ) was well established based on the developed soil shrinkage model (R 2 > 0.990, RMSE <0.012). Compared with the AFP under a constant pore assumption (AFP c ), the AFP sh was significantly decreased by 10.6% to 60.3% as soil water matric potentials were at <À33 kPa in the 0-10 cm layer or at <À10 kPa in the 10-20 cm layer (p < 0.05). According to the AFP sh curve of the Vertisol, soil moisture content (θ) was approximately 68%-80% of the field capacity when AFP sh reached non-limiting status (0.10 cm 3 cm À3 ), where the soil matric potential was close to the wilting point (À1500 kPa). Deep tillage treatment (SS and DP) increased AFP sh , showing good performance in decreasing the risk of soil aeration deficit relative to the NT and RT treatment. Our results demonstrate that neglecting the porosity variations during shrinkage may lead to a high bias in the AFP prediction.
Highlights• An equation of AFP associated with soil shrinkage was established.• AFP was significantly decreased when soil shrinkage was considered for a Vertisol.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.