86 results about "Arithmetic processing unit" patented technology

An example computing device may include memory and one or more processors. The one or more processors are configured to determine a Cartesian grid in a space. The Cartesian grid may correspond to a precision of an arithmetic processing unit. The one or more processors are configured to, for each of a plurality of neural network weights, determine a respective vector in the space, such that each respective vector belongs to a corresponding predefined vector on the Cartesian grid and represents a respective compressed neural network weight of a plurality of compressed neural network weights. The one or more processors are configured to execute, based on the plurality of compressed network weights, a neural network. The one or more processors are configured to generate, based on executing the neural network, an output.

The present disclosure discloses a product recognition apparatus and method, in which the apparatus includes: a vision sensor configured to obtain and transmit image information of a product to an arithmetic processing unit; the arithmetic processing unit configured to deal the image information of the product based on an image recognition model, to obtain and transmit category information of the product to a data transceiver; and the data transceiver configured to transmit the category information of the product to a backend server of a seller, so that the backend server of the seller makes a settlement based on the category information of the product and preset price information of different categories of products. By adopting the product recognition apparatus, it is possible to quickly recognize the categories of various products, increase efficiency and accuracy of the cashing, and improve the user experiences.

To improve accuracy to detect an abnormality of an electric motor.SOLUTION: An electric motor vibration model generation device 100 includes a current data preprocessing unit 104, a vibration data preprocessing unit 109, a time-series adjustment unit 110, and a first arithmetic processing unit 111. The current data preprocessing unit 104 extracts a plurality of kinds of feature amounts from input current data acquired from an electric motor to output them as preprocessing current data in association with an acquisition time of the input current data. The vibration data preprocessing unit 109 extracts a plurality of kinds of feature amounts from vibration data acquired from the electric motor to output them as preprocessing vibration data in association with an acquisition time of the vibration data. The time-series adjustment unit 110 adjusts a time series so that the preprocessing current data and the preprocessing vibration data are regarded as data acquired at the same time to output both of the data. The first arithmetic processing unit 111 generates a learning data set from the output of the time-series adjustment unit 110 to use the learning data set to train an electric motor vibration model to store the trained electric motor vibration model.SELECTED DRAWING: Figure 1

This oil evaluation device is provided with a calculation processing unit that calculates, on the basis of a measurement result of the density of oil in an internal combustion engine, the proportion of soot mixed into the oil as an evaluation index.

To provide a CMP apparatus and a polishing method that can easily improve the accuracy of measuring the thickness of a polishing layer among a plurality of layers stacked on the surface of a wafer.SOLUTION: There is provided a CMP apparatus 1 for polishing a polishing layer W1 of a wafer W comprising the polishing layer W1 and at least one underlayer W2 located below the polishing layer. The CMP apparatus comprises: a database 30 including respective model waveforms of reflected light spectra for respective thicknesses of the polishing layer corresponding to respective thicknesses of the underlayer; a detection unit 120 for determining the thickness of the polishing layer on the basis of the database and a first reflected light spectrum detected during polishing of the wafer; and an arithmetic processing unit 130 for generating the database on the basis of information on materials of the polishing layer, information on materials of the underlayer, a first thickness range of the polishing layer, a second thickness range of the underlayer, and thickness intervals of the polishing layer and the underlayer.SELECTED DRAWING: Figure 1

To provide a target commodity authorization program and a target commodity authorization device which analyze commodity data of display shelves for usage by a company manufacturing products for consumers.SOLUTION: A target commodity authorization program is a program for allowing a computer including a storage medium and an arithmetic processing unit to operate each of the following steps. The steps include: an image data storage step of storing image data of shop shelves obtained by each photographing means in the storage medium; a shelf contents grasping step of grasping shelf steps of shelves, and commodities arranged on the shelf steps; and a ratio calculation step of specifying a previously set target commodity among the grasped commodities, grasping the number of commodities of the target commodity, and calculating a ratio of the target commodity among all commodities on the shelves for a previously set commodity category or each of sub-categories being subordinate categories of the category.SELECTED DRAWING: Figure 1

A motion state monitoring system according to the present disclosure is configured to monitor a motion state at a target part of the body of a subject, and include: an arithmetic processing unit configured to generate a calculation result representing the motion state based on a detection result of a sensor configured to detect the motion state; an image processing unit configured to perform image processing on at least one of a photographed image obtained by photographing the motion state and a drawn image obtained by drawing the motion state; and a display unit configured to display the image-processed image and the calculation result in synchronization with each other.

According to the present invention, a computation processing device for a calculation system that comprises memory, storage, and the computation processing device comprises a computation unit that reads each key of a set of data for secret-shared key-value pairs from the memory and the storage, uses a pseudo-random function to calculate a hash value for each key, calculates a substitution on the basis of the resulting series of hash values, uses the substitution to construct a hash table from the set of data as extracted from the memory and the storage, and uses secret sharing to store the hash table at the memory and the storage.

An inclination of a weighing apparatus is automatically detected by the apparatus itself while preventing an increase in the number of components. In order to achieve the object described above, a weighing apparatus includes a weight sensor, a built-in weight to be loaded on the weight sensor, an adding / removing unit for adding / removing the built-in weight, a memory storing a theoretical value of the built-in weight, and an arithmetic processing unit, wherein the arithmetic processing unit includes an inclination angle computing unit configured to obtain an apparatus inclination angle from an arc-cosine of a weighing value of the built-in weight and the theoretical value of the built-in weight.

To provide a device, system, and method that achieve updatable, large-scale, and complex control, by reducing the load on a CPU for data receiving processing, even when some devices or applications are updated and unused data continues to be transmitted.SOLUTION: A control device 101 includes a receiving unit 105 that receives and processes data transmitted from an external apparatus, a received data holding unit 106 that holds the received data, a selected data holding unit 109 that holds only data to be used by an application of the received data, an address correspondence information generation unit 107 that generates an address relationship between the respective holding units from data access information of the application, a data extraction unit 108 that generates a control signal including whether each piece of data in the received data holding unit can be written into the selected data holding unit and a write address and gives the control signal to the selected data holding unit, and a data transfer determination unit 110 that outputs a command to transfer the data held in the selected data holding unit to a calculation processing unit.SELECTED DRAWING: Figure 1A

A tensor calculation device, a data processor, a tensor calculation method and a storage medium. The tensor calculation device comprises a receiving unit, a tensor processing unit and an arithmetic processing unit; the tensor processing unit is configured to perform a first tensor calculation on a first tensor to obtain a first intermediate tensor; the tensor processing unit is further configured to obtain a virtual tensor and perform a second tensor calculation on a second tensor according to the virtual tensor to obtain a second intermediate tensor, wherein the second intermediate tensor and the second tensor have the same shape, and the numerical values of the elements at all corresponding positions of the second intermediate tensor and the second tensor have the same preset proportional relationship; and the arithmetic processing unit is configured to perform arithmetic operation processing on the second intermediate tensor and the first intermediate tensor to obtain a result tensor. The tensor calculation device does not need to load data for performing the arithmetic operation, and there is no bandwidth competition phenomenon, thereby greatly improving the overall efficiency of the operator and effectively utilizing the powerful computing power of the tensor calculation device itself.

A counting scale includes a weighing pan, a weighing sensor connected to the weighing pan, a storage unit including a registered data table storing, as registered data, a unique identification number, an article name, and a unit weight with respect to one article, an arithmetic processing unit configured to urge placement of one piece of the target article on the weighing pan, calculate a unit weighed value of the one piece of the target article from the weighing sensor, narrow down registered data including a unit weight within a predetermined range with respect to the unit weighed value as candidate data from the registered data table, and obtain sorted candidate data by selecting and sorting the candidate data in ascending order of absolute value of deviation from the unit weighed value, and a display unit configured to display the sorted candidate data.

Monitoring the delta value, which is an indicator of directional risk, and buying and selling the Nikkei 225 mini when the delta value changes from 0 require constant monitoring and buying and selling by users, which places a heavy burden on users and makes it nearly impossible for individual investors in particular. [Solution] The option delta neutral automated system 3 calculates the delta value of an option based on stock price fluctuations and market price information acquired in real time from a market price information acquisition unit 31 connected to a market price information database 21 of a financial instruments trading device at a financial instruments business operator. When the delta value is not 0, an arithmetic processing unit 32 makes an automatic trading decision for the held futures in order to return the delta value to 0, and performs automatic trading processing for the held futures based on the decision of the arithmetic processing unit 32.

To provide an information processing apparatus, etc. capable of supporting construction of thermal environments in both a bedroom and a bed suitable for an individual user.SOLUTION: An information processing apparatus 100 includes a storage unit 130 and an arithmetic processing unit 120. The storage unit 130 stores attribute information including one or both of user's biological information and user's preference information about coldness and hotness. The arithmetic processing unit 120 specifies to which of a plurality of preset thermal sensation groups the thermal sensation of the user belongs, using user's attribute information stored in the storage unit 130, and determines a combination of a room temperature and warming performance of bedding and sleep wear while the user is sleeping according to the thermal sensation group that the user's thermal sensation belongs. The information processing apparatus 100 further includes an output unit 140 for outputting the combination of the room temperature and the warming performance of bedding and sleep wear while the user is sleeping.SELECTED DRAWING: Figure 1

The invention discloses a convolutional neural network accelerator based on a lookup table and in-memory computation, which comprises an on-chip main memory and a plurality of operation processing units, and is characterized in that each operation processing unit comprises a data preprocessing unit, an in-memory computation array unit, a first summator and an accumulator which are connected in sequence; serial input based on bit feature values is adopted, and an input feature value sequence is compressed; the data preprocessing unit is used for decompressing the compressed feature value sequence and the compressed weight, and matrix multiplication based on a lookup table form is carried out in the in-memory calculation array unit through an in-memory calculation macro module; the first adder is used for adding a first part and a result output by each column of in-memory computing macro-module in the in-memory computing array unit to obtain a second part and a result, and the accumulator is used for calculating the second part and the result to obtain a multiply-accumulate result corresponding to the characteristic value data stream; and high sparseness caused by unstructured pruning can be effectively utilized.

To provide an advertising banner production system which produces an advertising banner having a good advertising effect in a short time.SOLUTION: In an arithmetic processing unit 4 in an advertising banner production system, a foreground image production unit 11 produces a foreground image to be used in an advertising banner candidate, an advertising text production unit 12 produces an advertising text to be used in the advertising banner candidate, an advertising banner candidate production unit 13 combines the foreground image produced by the foreground image production unit 11 and the advertising text produced by the advertising text production unit 12 to thereby produce the advertising banner candidate, and a click rate prediction unit 14 predicts a click rate of the produced advertising banner candidate to be used.SELECTED DRAWING: Figure 1

SOLUTION: A computation apparatus comprises a plurality of memory banks and a computation processing unit that generates a result data sequence obtained by computation of a plurality of data sequences by repeating processes of reading the data sequence from a source memory bank among the plurality of memory banks, performing computation between the read data sequence and the next data sequence, and writing the data sequence obtained as a result of the computation into a destination memory bank among the plurality of memory banks while alternately switching the source memory bank and the destination memory bank.SELECTED DRAWING: Figure 4

According to one embodiment, the augmented reality presentation system comprises a haptic presentation unit which presents a haptic sensation to the specific part, a display unit, an information acquiring unit for acquiring spatial position information of the specific part, an information setting unit for setting haptic information corresponding to the haptic sensation presented to the specific part, and an arithmetic processing unit for generating a specific part image that simulates the specific part.

To provide a technique for displaying a change history of setting values and image data, etc., in association with time when a change instruction history is stored in a plurality of storage devices.SOLUTION: A PLC system comprises a PLC, an arithmetic processing unit, a storage unit, and a communication unit. The storage unit stores PLC program data, image data, and time chart data. The arithmetic processing unit comprises: a history management unit which stores a first change history as a change history of setting values accepted through a network connected by the communication unit and used for control of a target device, in the storage unit; a communication execution unit which acquires a second change history as a change history accepted via an operation panel and stored in the operation panel storage unit; and an image creation unit which creates list display image data for listing at least any of the first change history and the second change history, a PLC program, an image, and a time chart in association with time.SELECTED DRAWING: Figure 4

An embodiment is a computer including a plurality of accelerators, a computer for distributed processing includes a plurality of accelerators to each of which a part of a neural network is assigned and each of which is configured to derive a learning result based on input data and update each parameter value included in the part of the neural network by using the learning result; a plurality of network interface circuits each of which is configured to transmit and receive information on learning including the learning result via a network, and an arithmetic processing unit that is configured to control the plurality of accelerators and the plurality of network interface circuits to cause each of the plurality of accelerators to derive a learning result based on input data and to cause the plurality of network interface circuits to transmit and receive information on learning including the learning result.

A control unit (20) for controlling power conversion devices comprises a first arithmetic processing unit (21) and a second arithmetic processing unit (22) connected to each other via a transmission line (L1, L2). The first arithmetic processing unit (21) has a first calculator (31) that determines a basic torque command value, which is a common torque command value for motors arranged in a vehicle. The second arithmetic processing unit (22) has a second calculator (41) that adjusts the basic torque command value based on a parameter and thereby determines a shaft torque command value for each of the power conversion devices. This shaft torque command value is a torque command value for one or more motors that are to receive electrical power supplied by the power conversion device.The first arithmetic processing unit (21) includes an abnormality determiner (32) which determines the presence or absence of an abnormality in at least one of the transmission line (L1, L2) or the second computer (41) based on the base torque command value and the shaft torque command values.

An abnormal sound diagnosis system includes a sound acquisition unit configured to acquire data of a sound generated from an object, an inquiry information acquisition unit configured to acquire inquiry information regarding an abnormal sound generated in the object, an arithmetic processing unit configured to acquire a spectrogram indicating a relationship among a time, a frequency, and sound pressure from the data of the sound, an extraction unit configured to acquire, based on the inquiry information acquired by the inquiry information acquisition unit, an inferred frequency range of the abnormal sound generated in the object and extract an extracted range corresponding the inferred frequency range of the spectrogram acquired by the arithmetic processing unit, and a diagnosis unit configured to diagnose, based on the extracted range of the spectrogram extracted by the extraction unit, a cause of the abnormal sound generated in the object.

Provided are a device and a method which are capable of autonomously extending and changing the control of even a control device differing in instruction format and protocol. The present technology includes, for example, an extension control device that can communicably connect to a device having a control device, the extension control device including an arithmetic processing device, a memory in which a predetermined control program is stored, and a sensor, wherein the arithmetic processing device is configured to transmit the predetermined control program to the control device in response to the reception of a predetermined signal from the sensor.

To provide a power supply device that, even when output voltage output from a power supply voltage generation unit is unstable, can stabilize the output voltage to an arithmetic processing unit after a transition to a power saving mode.SOLUTION: When the operation mode of an image forming apparatus is switched to a power saving mode, output voltage output from a power supply circuit to a main control unit is stepped down to have a voltage value obtained by subtracting a correction value based on a prescribed step-down correction value (first step-down correction value, or second step-down correction value) from a reference voltage V0.SELECTED DRAWING: Figure 2

To provide a computer system with a small circuit area and lower power consumption.SOLUTION: A computer system includes a computer node having a processor and a NAND-type memory device with a three-dimensional structure. The NAND-type memory device with a three-dimensional structure has a first string and a second string, which belong to different blocks, respectively. The first string has a first memory cell, and the second string has a second memory cell. A controller writes first data to the first memory cell by receiving the first data and a signal including an instruction to write the first data. Subsequently, the controller reads the first data from the first memory cell and writes the first data to the second memory cell. This enables the computer node to be configured without providing a main memory such as DRAM.SELECTED DRAWING: Figure 1

To provide a semiconductor device which can improve arithmetic capacity such as AI learning and processing efficiency.SOLUTION: A flash memory 100 includes a NAND type or NOR type memory cell array 110, and an arithmetic processing unit 190. The arithmetic processing unit 190 includes a bit line current detection unit 200, a voltage holding unit 210 which holds voltage corresponding to detected current, an addition unit 220 which adds the voltage held in the voltage holding unit 210, and an A / D conversion unit 230 which performs A / D conversion on an addition result of the addition unit 220. The arithmetic processing unit 190 can calculate the sum of currents which flow through a bit line in a line direction and / or a column direction at reading of the memory cell array.SELECTED DRAWING: Figure 2

In the arithmetic processing device, the arithmetic control unit controls in the following manner: during the filtering process and the accumulation process for a specific pixel used to calculate the output feature quantity map, the intermediate result is temporarily stored in the memory for storing the accumulation result to perform the processing of other pixels. After the intermediate results of the accumulation process for all pixels are stored in the memory for storing the accumulation result, it returns to the first pixel, reads out the value stored in the memory for storing the accumulation result, uses this value as the initial value of the accumulation process, and continues to execute the accumulation process.

To appropriately manage the quality of a product when using a raw material having variations in characteristics or the like in many cases.SOLUTION: A control device 100 that operates an electrical device 120 constituting a manufacturing line of a product and acquires data from a measurement device 130 installed in the manufacturing line, the control device 100 including a controller unit 110 that operates the electrical device 120, an arithmetic processing unit 104 that has a function of calculating a parameter for controlling the electrical device 120 and outputting the parameter to the controller unit 110, a function of performing arithmetic processing on data acquired from at least one of the electrical device 120 and the measurement device 130, and a function of acquiring at least one of procurement information and storage information of a raw material of the product, 130 120, and a storage part 106 for storing an arithmetic processing result of the arithmetic processing part 104, wherein the arithmetic processing result is associated with at least one of the procurement information and the storage information.SELECTED DRAWING: Figure 1

A maintenance assistance system assists maintenance in a component mounting system including a plurality of work units. The maintenance assistance system includes: a display unit that displays a maintenance event of a work unit that needs to be maintained among the plurality of work units; a display control unit that controls the display unit; and an arithmetic processing unit that performs processing for obtaining data of a parameter indicating the work accuracy of the work unit per unit period, and processing for detecting the work unit, the work accuracy of which is reduced, as a work unit to be maintained on the basis of the data. The display control unit controls a display method and / or display content of maintenance items of the target work unit such that maintenance of the target work unit is prioritized than maintenance of other work units.

To efficiently use resources in FPGA by eliminating auxiliary processing circuits that are provided in an FPGA circuit in preparation for software error and do not operate in normal processing.SOLUTION: A data processing system includes a computing unit comprising a programmable device, and an arithmetic processing unit connected to the computing unit. The computing unit includes: a plurality of processing circuits that executes, in parallel, data processing instructions provided from the arithmetic processing unit; an error detection unit which detects software error generated in a processing circuit executing a data processing instruction; and a processing circuit selection unit which selects a processing circuit to execute the data processing instruction from among the plurality of processing circuits. The processing circuit selection unit identifies a processing circuit in which the software error occurred, on the basis of the software error detected by the error detection unit, and selects a processing circuit to execute the data processing instruction, excluding the processing circuit in which the software error occurred, out of the plurality of processing circuits.SELECTED DRAWING: Figure 1