Tongji University

Advanced Micro Devices Corporation

Intel Corporation

S. C. Johnson Son, Inc.

Other

Cisco Technology, Inc.

East China University of Science and Technology

Kohler (china) Investment Co., Ltd

Gracell Biotechnologies (shanghai) Co., Ltd.

Kewpie Corporation

Fbc (shanghai) Pharmaceutical Technology Co., Ltd.

Hua Zhang

Yang Liu

Fan Lyu

Pinjing He

Liming Shao

Min Yu

Song Zhang

Jiantan Liu

Jiajia Li

Wei Cao

Yan Li

Hao Liu

Ying Luo

Jie He

Vikram Kumaran

Liping Liu

Lu Liu

Junhua Hou

Huan Shi

Weihan Wang

Zhi Hong Yu

Sijia Wang

Zhitao Shen

Xinhua Zhong

David Tang

Xiaomin Chen

Wei Peng

Zhan Yang

Xiangning Ma

Changlu Zheng

Yifei Zheng

Xin Kang

Li Gan

Lianjun Shen

Yanan Wang

Jiqin Zhang

Jianqing Zhang

Junjie Qiu

Jinlong Yang

The present invention relates to a macro plastic and micro plastic detection method based on RGB and hyperspectral image fusion, which includes the following steps: obtaining macro plastics and micro plastics; mixing with solid wastes to obtain a solid-phase substrate; pretreating the obtained solid-phase substrate to obtain a material; drying to remove part of moisture and coating on a quartz window sheet, drying until the moisture is completely removed, and flattening by using another quartz window sheet to obtain a material to be detected; obtaining an RGB image and a hyperspectral image of the material to be detected respectively by using a high-resolution color image scanner and a hyperspectral camera; fusing the obtained RGB image and hyperspectral image; and automatically classifying and identifying the macro plastics and the micro plastics by using a supervised classification model.

Macro plastic and micro plastic detection method based on RGB and hyperspectral image fusion

Techniques related to viewport selection in immersive video contexts are discussed. Such techniques include generating multiple viewport predictions each for a future time interval and based on different prediction models, ranking the viewport predictions using error descriptors of the prediction models, selecting a viewport prediction for the future time intervals using the ranking, and correcting the selected viewport predictions using the error descriptors.

Real-time adaptive correction in viewport prediction for improved immersive video

A bathtub system includes a bathtub body and a bath plate. The bath tub body defines a bathing volume. The bath plate is configured to detachably connect with the bathtub body within the bathing volume. The bath plate includes a bath plate body and telescopic pieces arranged on two sides of the bath plate body. The telescopic pieces are fixedly connected with the bath plate body. A wall of the bathtub body includes a wall protrusion configured to retain the telescopic piece.

Bathtub bath board and bathtub

The invention relates to the field of coal mine electric locomotives, and particularly discloses an autonomous positioning method for an underground coal mine anti-explosion storage battery rail electric locomotive, which comprises the following steps: constructing an underground coal mine dynamic map, scanning the surrounding environment by a laser radar at the frequency of more than 10 HZ, and transmitting data to a vehicle-mounted controller; performing dynamic map construction by the vehicle-mounted controller according to point cloud distribution map; underground electric locomotive is accurately positioned by adopting a static accurate positioning mode, a dynamic accurate positioning mode and a multi-track lateral positioning mode; through fusion of multiple dynamic positioning modes, laser radar positioning is adopted, inertial positioning and passive identification card RF ID positioning are fused, and the positioning precision of the system is greatly improved; three positioning modes are organically fused to provide system adaptability, stability and reliability.

Autonomous positioning method for underground coal mine anti-explosion storage battery rail electric locomotive

An apparatus and method are described for graphics virtualization with late synchronization. For example, a virtualized graphics execution environment is implemented on a host, including a hypervisor to virtualize graphics processing resources for one or more virtual machines (VMs). A plurality of graphics commands are received responsive to execution of a graphics application in one of the VMs. The graphics commands are persisted until a frame triggering event is detected. Upon detection of a frame triggering event, the persisted graphics commands are submitted to the host.

Apparatus and method for graphics virtualization with late synchronization

A method for removing calcium ions from high concentration organic wastewater is provided. The method comprises the steps of: (1) introducing high concentration organic wastewater containing Ca, inorganic carbon and a seed crystal into a reactor with a molar ratio of Cato inorganic carbon of 1:(3.2-6.2); (2) adjusting the hydrogen ion activity α(H) and ionic strength of the solution in the reactor; (3) sequentially stirring and precipitating in the reactor to convert Cain the high concentration organic wastewater into calcium carbonate which is then precipitated for calcium removal.

Method for removing calcium ions from high concentration organic wastewater

Systems, apparatuses, and methods for converting data to a tiling format when implementing convolutional neural networks are disclosed. A system includes at least a memory, a cache, a processor, and a plurality of compute units. The memory stores a first buffer and a second buffer in a linear format, where the first buffer stores convolutional filter data and the second buffer stores image data. The processor converts the first and second buffers from the linear format to third and fourth buffers, respectively, in a tiling format. The plurality of compute units load the tiling-formatted data from the third and fourth buffers in memory to the cache and then perform a convolutional filter operation on the tiling-formatted data. The system generates a classification of a first dataset based on a result of the convolutional filter operation.

Tiling format for convolutional neural networks

The present invention provides an engineered immune cell targeting BCMA and use thereof. In particular, the present invention provides a CAR specifically targeting BCMA, the CAR comprising an antigen-binding domain which is an S-derived scFv, an antibody heavy chain variable region as shown in SEQ ID NO: 9 and an antibody light chain variable region as shown in SEQ ID NO: 10. The present invention also provides a CAR-T cell comprising the CAR, a double CAR- and CAR-T cell comprising the S-derived scFv, and related use thereof. Compared to CAR-T cells constructed using other scFvs, the constructed CAR-T cell of the present invention has a better killing effect and tumor elimination capability.

Engineered immune cells targeting BCMA and their uses thereof

Systems, apparatuses, and methods for implementing a tiling algorithm for a matrix math instruction set are disclosed. A system includes at least a memory, a cache, a processor, and a plurality of compute units. The memory stores a plurality of matrix elements in a linear format, and the processor converts the plurality of matrix elements from the linear format to a tiling format. Each compute unit retrieves a plurality of matrix elements from the memory into the cache. Each compute unit includes a matrix operations unit which loads the plurality of matrix elements of corresponding tile(s) from the cache and performs a matrix operation on the plurality of matrix elements to generate a result in the tiling format. The system generates a classification of a first dataset based on results of the matrix operations.

Tiling algorithm for a matrix math instruction set

The present invention relates to a self-deodorizing composting system, comprising an outer membrane, an inner membrane, an air-ribbed cage and a deodorizing base, wherein two deodorizing bases are provided; the outer membrane and the inner membrane are held by the air-ribbed cage, with both ends of the outer membrane and the inner membrane respectively fixed to the two deodorizing bases, whereby a closed space is formed between the outer membrane and the inner membrane; the inner side of the inner membrane is used for placing the composting piles, with the malodor and water vapor generated from the composting piles entering into the closed space formed between the outer membrane and the inner membrane through the inner membrane, so that the water vapor is condensed into the water in the closed space, and the malodor molecules are absorbed and dissolved into the condensed water, thus forming the waste water which falls along the inner side of the outer membrane and the outer side of the inner membrane to the bottom of the deodorizing bases and is discharged from the deodorizing base. Comparing with the prior art, the system in the present invention has a simple and detachable structure, and can isolate and remove the malodor and the water vapor generated when the biomass waste is composted.

Self-deodorizing composting system

The present invention relates to a method for inhibiting the formation of dioxin-like compounds in solid waste incineration flue gases, belonging to the field of cleaning of solid waste incineration flue gases. In accordance with the present invention, when incineration flue gases cool down to 500&#x2dc;450&#xb0; C., the flue gases are introduced into an inhibition reactor where copper chlorides in flue gas particulates mix and react with inhibitors to convert into copper metaphosphate so as to inactivate the copper chlorides which can catalyze the formation of dioxin-like compounds and control dioxin-like compound pollutants in incineration flue gases at the source. Compared with the prior art, the invention can effectively control the main formation ways of dioxin-like compounds in solid waste incineration flue gases by optimizing inhibitors and reaction conditions. The method of the invention does not affect the residual heat utilization of solid waste incineration flue gases, so solid waste incineration has a better resource utilization effect. The ammonium dihydrogen phosphate inhibitor used in the invention has the advantages of high inhibition efficiency, strong operability, low cost and environment protection, providing the technology with good application feasibility.

Method of inhibiting formation of dioxins and dioxin-like compounds in solid waste incineration flue gas

The present invention discloses a method and a system of treating biomass wastes by biochemistry-thermochemistry multi-point interconnection. The present invention applies solid, gas and liquid products of the thermochemical treatment subsystem to the biochemical treatment subsystem and applies heat produced by the biochemical treatment subsystem to the thermochemical treatment subsystem, forming multi-point and two-way interconnection between the biochemical treatment subsystem and the thermochemical treatment subsystem, thereby increasing the yield and stability of energy gas of the biochemical treatment subsystem and reducing pollution and energy consumption of the thermochemical treatment subsystem respectively. The present invention is suitable for treating biomass wastes with high and low water contents at the same time, producing soil amendment, liquid fuel and biogas, having properties of low secondary pollution and significant reduction of greenhouse gas emission and so on. The bio-stability, humus content and nitrogen content of the solid product are as high as soil amendment, making it easy to store and transport.

Method and system of treating biomass wastes by biochemistry-thermochemistry multi-point interconnection

Location data is obtained from signals transmitted by a first plurality of mobile wireless devices in a wireless network, wherein the first plurality of mobile wireless devices are moving within a predefined space, and wherein the location data comprises a plurality of location data time points, each location data time point including a timestamp, a unique mobile wireless device identifier, and location information indicating where in the predefined space an associated mobile wireless device is located. For each mobile wireless device, location data time points are aggregated to generate a set of aggregated location data for each mobile wireless device, and the set of aggregated location data is analyzed to determine characteristics corresponding to time-dependent behavior and location-specific behavior of the corresponding mobile wireless device. A user of each corresponding mobile wireless device is classified into a category of a plurality of categories based on the determined characteristics.

Leveraging location data from mobile devices for user classification

Growing community of inventors

Shanghai, China

Hua Zhang