Abstract: This disclosure addresses the single-image compressive sensing (CS) and reconstruction problem. A scalable Laplacian pyramid reconstructive adversarial network (LAPRAN) facilitates high-fidelity, flexible and fast CS image reconstruction. LAPRAN progressively reconstructs an image following the concept of the Laplacian pyramid through multiple stages of reconstructive adversarial networks (RANs). At each pyramid level, CS measurements are fused with a contextual latent vector to generate a high-frequency image residual. Consequently, LAPRAN can produce hierarchies of reconstructed images and each with an incremental resolution and improved quality. The scalable pyramid structure of LAPRAN enables high-fidelity CS reconstruction with a flexible resolution that is adaptive to a wide range of compression ratios (CRs), which is infeasible with existing methods.

Abstract: A compression ratio (CR) adapter (CRA) for end-to-end data-driven compressive sensing (CS) reconstruction (EDCSR) frameworks is provided. EDCSR frameworks achieve state-of-the-art reconstruction performance in terms of reconstruction speed and accuracy for images and other signals. However, existing EDCSR frameworks cannot adapt to a variable CR. For applications that desire a variable CR, existing EDCSR frameworks must be trained from scratch at each CR, which is computationally costly and time-consuming. Embodiments described herein present a CRA framework that addresses the variable CR problem generally for existing and future EDCSR frameworks with no modification to given reconstruction models nor enormous additional rounds of training needed. The CRA exploits an initial reconstruction network to generate an initial estimate of reconstruction results based on a small portion of acquired image measurements.

Abstract: A data-driven nonuniform subsampling approach for computation-free on-sensor data dimensionality is provided, referred to herein as selective sensing. Designing an on-sensor data dimensionality reduction scheme for efficient signal sensing has long been a challenging task. Compressive sensing is a generic solution for sensing signals in a compressed format. Although compressive sensing can be directly implemented in the analog domain for specific types of signals, many application scenarios require implementation of data compression in the digital domain. However, the computational complexity involved in digital-domain compressive sensing limits its practical application, especially in resource-constrained sensor devices or high-data-rate sensor devices. Embodiments described herein provide a selective sensing framework that adopts a novel concept of data-driven nonuniform subsampling to reduce the dimensionality of acquired signals while retaining the information of interest in a computation-free fashion.

Abstract: A method for predicting mtDNA mutation threshold, fertility risk and oocyte retrieval number includes: establishing an incidence probability prediction model of mitochondrial mutation and estimating a mutation threshold s; establishing a fertility risk prediction model and predicting a fertility risk of a mutation carrier; and establishing an oocyte retrieval prediction model by binomial distribution and calculating oocyte retrieval number by PGT of a mutation carrier. The incidence probability prediction model is used to estimate a threshold value of common mtDNA mutations and predict a mother's fertility risk and the oocyte retrieval number by PGT. On the basis of the fertility risk prediction model and oocyte retrieval prediction model, the fertility risk and a minimum oocyte retrieval number by PGT needed to give birth to a healthy offspring can be calculated by knowing a mitochondrial DNA mutation ratio of mutation carrier which is helpful for genetic management and genetic consultation.

Abstract: Systems and methods are provided for detecting and correcting synchronization errors in multimedia content comprising a video stream and a non-original audio stream. Techniques for directly detecting synchronization of video and audio streams may be inadequate to detect synchronize errors for non-original audio streams, particularly where such non-original audio streams contain audio not reflective of events within the video stream, such as speaking dialog in a different language than the speakers of the video stream. To overcome this problem, the present disclosure enables synchronization of a non-original audio stream to another audio stream, such as an original audio stream, that is synchronized to the video stream. By comparison of signatures, the non-original and other audio stream are aligned to determine an offset that can be used to synchronize the non-original audio stream to the video stream.

Abstract: This disclosure addresses the single-image compressive sensing (CS) and reconstruction problem. A scalable Laplacian pyramid reconstructive adversarial network (LAPRAN) facilitates high-fidelity, flexible and fast CS image reconstruction. LAPRAN progressively reconstructs an image following the concept of the Laplacian pyramid through multiple stages of reconstructive adversarial networks (RANs). At each pyramid level, CS measurements are fused with a contextual latent vector to generate a high-frequency image residual. Consequently, LAPRAN can produce hierarchies of reconstructed images and each with an incremental resolution and improved quality. The scalable pyramid structure of LAPRAN enables high-fidelity CS reconstruction with a flexible resolution that is adaptive to a wide range of compression ratios (CRs), which is infeasible with existing methods.

Abstract: This disclosure addresses the single-image compressive sensing (CS) and reconstruction problem. A scalable Laplacian pyramid reconstructive adversarial network (LAPRAN) facilitates high-fidelity, flexible and fast CS image reconstruction. LAPRAN progressively reconstructs an image following the concept of the Laplacian pyramid through multiple stages of reconstructive adversarial networks (RANs). At each pyramid level, CS measurements are fused with a contextual latent vector to generate a high-frequency image residual. Consequently, LAPRAN can produce hierarchies of reconstructed images and each with an incremental resolution and improved quality. The scalable pyramid structure of LAPRAN enables high-fidelity CS reconstruction with a flexible resolution that is adaptive to a wide range of compression ratios (CRs), which is infeasible with existing methods.

Abstract: A data-driven nonuniform subsampling approach for computation-free on-sensor data dimensionality is provided, referred to herein as selective sensing. Designing an on-sensor data dimensionality reduction scheme for efficient signal sensing has long been a challenging task. Compressive sensing is a generic solution for sensing signals in a compressed format. Although compressive sensing can be directly implemented in the analog domain for specific types of signals, many application scenarios require implementation of data compression in the digital domain. However, the computational complexity involved in digital-domain compressive sensing limits its practical application, especially in resource-constrained sensor devices or high-data-rate sensor devices. Embodiments described herein provide a selective sensing framework that adopts a novel concept of data-driven nonuniform subsampling to reduce the dimensionality of acquired signals while retaining the information of interest in a computation-free fashion.

Abstract: A compression ratio (CR) adapter (CRA) for end-to-end data-driven compressive sensing (CS) reconstruction (EDCSR) frameworks is provided. EDCSR frameworks achieve state-of-the-art reconstruction performance in terms of reconstruction speed and accuracy for images and other signals. However, existing EDCSR frameworks cannot adapt to a variable CR. For applications that desire a variable CR, existing EDCSR frameworks must be trained from scratch at each CR, which is computationally costly and time-consuming. Embodiments described herein present a CRA framework that addresses the variable CR problem generally for existing and future EDCSR frameworks with no modification to given reconstruction models nor enormous additional rounds of training needed. The CRA exploits an initial reconstruction network to generate an initial estimate of reconstruction results based on a small portion of acquired image measurements.

Abstract: An anti-agglomeration device for a nanofluid includes a support module, a motion module, a photoacoustic conversion module and a control module. The support module includes a frame and screws and is configured to support the photoacoustic conversion module and the motion module. The photoacoustic conversion module includes a nanosecond laser, a first clamp, a lens, a silica optical fiber, a second clamp, gold nanoparticles and a container and is configured to realize photoacoustic conversion to generate ultrasonic waves. The motion module includes a servo motor, a dovetailed rail, a guide screw, a fixed plate, a slider and a deep groove ball bearing. The motion module is configured to support the photoacoustic conversion module and realize the combined motions of the photoacoustic conversion module. The control module includes a support plate and a CCD camera and is configured to control the motion module in real time.

Abstract: An anti-agglomeration device for a nanofluid includes a support module, a motion module, a photoacoustic conversion module and a control module. The support module includes a frame and screws and is configured to support the photoacoustic conversion module and the motion module. The photoacoustic conversion module includes a nanosecond laser, a first clamp, a lens, a silica optical fiber, a second clamp, gold nanoparticles and a container and is configured to realize photoacoustic conversion to generate ultrasonic waves. The motion module includes a servo motor, a dovetailed rail, a guide screw, a fixed plate, a slider and a deep groove ball bearing. The motion module is configured to support the photoacoustic conversion module and realize the combined motions of the photoacoustic conversion module. The control module includes a support plate and a CCD camera and is configured to control the motion module in real time.

Abstract: The present application discloses a single-point diamond dresser for a grinding wheel based on acoustic emission online monitoring, which includes a support module, an anti-interference module, a compression cooling module and an acoustic emission online monitoring module. An acoustic emission sensor monitors the dressing state of the grinding wheel online; a damping sheet and a damping interlayer greatly reduce external noise interference; a high pressure coolant causes an upward elastic deformation of an elastic spacer and the damping sheet, enlarging a contact force between the acoustic emission sensor and the core and moreover improving a sensitivity of the acoustic emission sensor; the coolant flows through the coolant passages in the core to cool a dressing area; current limiting passages, pressure relief cavities and perforated pressure-relief plates limit flow and reduce a pressure of the coolant, reducing the interference of the coolant on the grinding wheel dressing.

Abstract: This disclosure addresses the single-image compressive sensing (CS) and reconstruction problem. A scalable Laplacian pyramid reconstructive adversarial network (LAPRAN) facilitates high-fidelity, flexible and fast CS image reconstruction. LAPRAN progressively reconstructs an image following the concept of the Laplacian pyramid through multiple stages of reconstructive adversarial networks (RANs). At each pyramid level, CS measurements are fused with a contextual latent vector to generate a high-frequency image residual. Consequently, LAPRAN can produce hierarchies of reconstructed images and each with an incremental resolution and improved quality. The scalable pyramid structure of LAPRAN enables high-fidelity CS reconstruction with a flexible resolution that is adaptive to a wide range of compression ratios (CRs), which is infeasible with existing methods.

Abstract: The present application discloses a single-point diamond dresser for a grinding wheel based on acoustic emission online monitoring, which includes a support module, an anti-interference module, a compression cooling module and an acoustic emission online monitoring module. An acoustic emission sensor monitors the dressing state of the grinding wheel online; a damping sheet and a damping interlayer greatly reduce external noise interference; a high pressure coolant causes an upward elastic deformation of an elastic spacer and the damping sheet, enlarging a contact force between the acoustic emission sensor and the core and moreover improving a sensitivity of the acoustic emission sensor; the coolant flows through the coolant passages in the core to cool a dressing area; current limiting passages, pressure relief cavities and perforated pressure-relief plates limit flow and reduce a pressure of the coolant, reducing the interference of the coolant on the grinding wheel dressing.

Abstract: A wave power generation device, including a power generation assembly, a mooring system, a floating platform, floating towers, control cabins, a connecting member, and anti-sway plates. The floating platform is a square floating box. The power generation assembly includes a swing plate, a hydraulic cylinder, an energy accumulator, a hydraulic motor, a generator, a battery, and a power generation and distribution device. The floating platform includes a main deck. The main deck includes a manhole and a support for supporting the swing plate. The swing plate is connected to the support via a first hinge. One end of the hydraulic cylinder is connected to the swing plate, and another end of the hydraulic cylinder is connected to a floating platform. The mooring system includes four anchor windlasses. A method for operating and maintaining the wave power generation device is also provided.

Abstract: An underwater towing device, including an underwater towing assembly and a fixing mechanism of the underwater towing assembly. The underwater towing assembly includes a first floating body, a second floating body, and a third floating body. The fixing mechanism includes a fairlead carrier and two supports disposed on two sides of the fairlead carrier. The first floating body, the second floating body, and the third floating body are connected together via a fixing frame. The first floating body, the second floating body, and the third floating body have streamlined and bilaterally symmetrical shapes and all are hollow to accommodate experimental equipment. The first floating body is a cylinder having a drop-shaped horizontal cross-section. The upper part of the first floating body is provided with a buoyancy reserve tank having an elliptic horizontal cross-section.

Abstract: A wave power generation device, including a power generation assembly, a mooring system, a floating platform, floating towers, control cabins, a connecting member, and anti-sway plates. The floating platform is a square floating box. The power generation assembly includes a swing plate, a hydraulic cylinder, an energy accumulator, a hydraulic motor, a generator, a battery, and a power generation and distribution device. The floating platform includes a main deck. The main deck includes a manhole and a support for supporting the swing plate. The swing plate is connected to the support via a first hinge. One end of the hydraulic cylinder is connected to the swing plate, and another end of the hydraulic cylinder is connected to a floating platform. The mooring system includes four anchor windlasses. A method for operating and maintaining the wave power generation device is also provided.

Abstract: An underwater towing device, including an underwater towing assembly and a fixing mechanism of the underwater towing assembly. The underwater towing assembly includes a first floating body, a second floating body, and a third floating body. The fixing mechanism includes a fairlead carrier and two supports disposed on two sides of the fairlead carrier. The first floating body, the second floating body, and the third floating body are connected together via a fixing frame. The first floating body, the second floating body, and the third floating body have streamlined and bilaterally symmetrical shapes and all are hollow to accommodate experimental equipment. The first floating body is a cylinder having a drop-shaped horizontal cross-section. The upper part of the first floating body is provided with a buoyancy reserve tank having an elliptic horizontal cross-section.

Abstract: An apparatus for weighing materials online is disclosed, which belongs to an art of online weighing of material. The apparatus comprises: a light-emitting unit, emitting light beams irradiating on the surface of materials transported on a belt to form a bright projection which has the same shape as that of a upper contour of a cross-section of the material; and a CCD camera for continuously picking up images of the bright projection on the upper contour of the cross-section of the material; and an image capture unit, connected to the CCD camera for continuously capturing the images; and a central processing unit, connected to the image capture unit for processing the images captured by the image capture unit to compute a weight of the material. The present invention has many advantages, such as, high precision, good reliability and stability, which applies to the belt without a constant speed and a condition that a bulk shape and the bulk density of the material change largely.

Abstract: An apparatus for weighing materials online is disclosed, which belongs to an art of online weighing of material. The apparatus comprises: a light-emitting unit, emitting light beams irradiating on the surface of materials transported on a belt to form a bright projection which has the same shape as that of a upper contour of a cross-section of the material; and a CCD camera for continuously picking up images of the bright projection on the upper contour of the cross-section of the material; and an image capture unit, connected to the CCD camera for continuously capturing the images; and a central processing unit, connected to the image capture unit for processing the images captured by the image capture unit to compute a weight of the material. The present invention has many advantages, such as, high precision, good reliability and stability, which applies to the belt without a constant speed and a condition that a bulk shape and the bulk density of the material change largely.