Fujitsu Corporation

Takanori Nakao

Tomonori Kubota

Yasuyuki Murata

Yoichi Koyanagi

Takeshi Miyamae

Yukihiko Hirayanagi

An image processing system includes one or more memories; and one or more processors coupled to the one or more memories and the one or more processors configured to acquire a degree of influence of each region of image data on a recognition result when recognition processing is performed for the image data, and reduce data size of the image data based on the degree of influence.

Image processing system, image processing method, and storage medium

A server includes: a memory; and a processor coupled to the memory and configures to: receive a frame image of a moving image compressed by an information processing device; detect a sign of deterioration in analysis accuracy based on accuracy information that represents certainty of an analysis result by artificial intelligence for the frame image; transmit, to the information processing device, instruction information of controlling a compression rate in compression of the moving image by the information processing device between an upper bound and a lower bound; and change the upper bound and the lower bound according to stability of the analysis accuracy based on the detection of a sign of deterioration in the analysis accuracy.

Server, control method, and computer-readable recording medium storing control program

A method includes: generating a refine image having a maximized correct label score of inference from an incorrect image by which an incorrect label is inferred by a neural network; generating a third map by superimposing a first map and a second map, the first map indicating pixels to each of which a change is made in generating the refine image, of plural pixels of the incorrect image, the second map indicating a degree of attention for each local region in the refine image, the each local region being a region that has drawn attention at the time of inference by the neural network, and the third map indicating a degree of importance for each pixel for inferring a correct label; and specifying an image section based on a pixel value of the third map, the image section corresponding to a region causing incorrect inference in the incorrect image.

Analysis apparatus, non-transitory computer-readable storage medium for analysis program, and analysis method

A method includes: generating a refine image from an incorrect image from which an incorrect label is inferred by a neural network; generating a third map by superimposing a first map and a second map, the first map indicating pixels to each of which a change is made in generating the refine image, of plural pixels in the incorrect image, the second map indicating a degree of attention for each local region in the refine image, each local region being a region that has drawn attention at the time of inference by the neural network, and the third map indicating a degree of importance for each pixel for inferring a correct label; and obtaining an added value for respective divided region in the third map by summing pixel values within the respective divided region, the respective divided region being a region divided according to a predetermined index.

Analysis apparatus, non-transitory computer-readable storage medium for storing analysis program, and analysis method

A storage control device configured to operate as one of a plurality of storage control devices in a distributed storage system configured to distributively arrange divided data in at least a part of the plurality of storage control devices, the divided data being created by dividing original data into a plurality of pieces. The storage control device is configured to: identify a storage position of management information to be used for reproducing of the original data; acquire the management information; store the management information in a first area corresponding to the position of the management information in the original data; store the divided data in a second area corresponding to a relative position of the divided data in the original data; and create reproduction data by using the divided data, the reproduction data including dummy data stored in a third area other than the first and second area.

Storage control device, distributed storage system, and non-transitory computer-readable storage medium for storing storage control program

An analysis method executed by a computer includes acquiring a refine image that maximizes a score for inferring a correct label by an inferring process using a trained model, the refine image being generated from an input image used when an incorrect label is inferred; generating a map indicating a region of pixels having the same or similar level of attention degree related to inference in the inferring process, of a plurality of pixels in the generated refine image, based on a feature amount used in the inferring process; extracting an image corresponding to a pixel region whose level in the generated map is a predetermined level, from calculated images calculated based on the input image and the refine image; and generating an output image that specifies a portion related to an inference error in the inferring process, among the calculated images, based on image processing on the extracted image.

Analysis apparatus, storage medium storing analysis program, and analysis method

A method includes: executing a reception process that includes receiving a request including any of a first request and a second request; and executing a control process that includes performing a first process when the first request is received, the first request being a request for executing a first migration process configured to migrate data from a first storage device to a second storage device having a higher access speed than the first storage device, the first process including recording state information and starting the execution of the first migration process, and performing a second process when the second request is received, the second request being a request for executing a second migration process configured to migrate the data from the second storage device to the first storage device, the second process including stopping the first migration process before starting the execution of the second migration process.

Information processing apparatus, method, and storage medium for avoiding accidental data deletion during data migration

A data compression device includes a processor configured to extract delimiter characters and character strings from compression target data. The character strings each are sandwiched between two of the delimiter characters in the compression target data. The processor is configured to generate first data by arranging the delimiter characters to allow an order in which the delimiter characters occur in the compression target data to be identified. The processor is configured to generate second data by arranging the character strings for each type of combination of two of the delimiter characters, which sandwich each of the character strings in the compression target data, to allow an order in which the character strings occur in the compression target data to be identified. The processor is configured to perform compression on the first data and the second data. The processor is configured to output a result of the compression.

Data compression device and data decompression device

A storage apparatus includes a processor. The processor is configured to sequence a plurality of data pieces. The plurality of data pieces are respectively stored in a plurality of memory devices. The processor is configured to set compensation ranges to be respectively compensated by a first predetermined number of parities. The compensation ranges are respective portions of consecutive data pieces among the sequenced data pieces. The compensation ranges include a variably set number of data pieces for the respective parities. Each of the plurality of data pieces is included in a second predetermined number of compensation ranges.

Storage system and storage apparatus

A parity-layout generating method, includes: creating a first local parity layout with a first calculation range for calculating local parity; creating a second local parity layout with a second calculation range for calculating local parity, a length of the second calculation range being different from a length of the first calculation range; and creating, by a computer, a third local parity layout with the first calculation range and the second calculation range by combining the first local parity layout and the second local parity layout.

Parity-layout generating method, parity-layout generating apparatus, and storage system

An n-tap (n&#x2267;2 (in the case of FIG., n=2)) DFE equalizes a received signal on the basis of a determination result of a signal which has been received, and outputs a determination result of the received signal after equalization. An adjustment section generates a plurality of filter patterns for observation of an inter-symbol interference amount by changing values of n bits included in the plurality of filter patterns. The adjustment section then compares each of the plurality of filter patterns generated with an output data string corresponding to the determination result of the received signal after equalization and adjusts tap coefficients set for n (in the case of FIG.) taps included in the DFE on the basis of adjustment values corresponding to values of the n bits included in a filter pattern which matches the output data string.

Receiver circuit

A circuit includes: a first adder configured to add a first offset cancellation value to an input signal value; a second adder configured to add a first equalization value to an output signal value from the first adder; a first comparator configured to make a binary decision on an output signal value from the second adder; a third adder configured to add a second offset cancellation value to the input signal value; a fourth adder configured to add a second equalization value to an output signal value from the third adder; a second comparator configured to make a binary decision on an output signal value from the fourth adder; a selector configured to output a determination result of the first comparator or a determination result of the second comparator in accordance with a determination result of preceding one bit of the input signal value.

Growing community of inventors

Kawasaki, Japan

Takanori Nakao