7 results about "Path design" patented technology

The present invention provides a method for laser rotary cutting of a ceramic substrate, which relates to the technical field of laser cutting. The method includes designing a laser rotary cutting path drawing in path design software according to the thickness of the ceramic substrate and the cutting profile dimensions; converting the laser rotary cutting path drawing by using the drawing processing software supporting the laser device; downloading the converted laser rotary cutting path drawing to the laser device and adjusting the parameters of the laser device; and starting the laser device to cut the ceramic substrate. Through a more refined energy action mode, the distribution of laser energy on the material is made more uniform, thereby reducing the risk of local overheating and reducing the generation of the heat affected zone and slag. The laser rotary cutting path is designed in advance, which improves the cutting efficiency while ensuring the cutting quality, and helps to obtain a more vertical and flat cutting edge, thereby improving the dimensional accuracy of the cut piece.

Embodiments of the present disclosure disclose a method, a processor, an apparatus, a storage medium and a product for quick access of a cache. The method comprises: dividing a static random access memory array of a target cache module into two regions according to physical distances from an upstream device; a region closer to the upstream device is a proximal region, and a region farther from the upstream device is a distal region; constructing a first pipeline transmission path between the upstream device and the proximal region; constructing a second pipeline transmission path between the upstream device and the distal region; the first pipeline transmission path contains fewer pipeline register stages than the second pipeline transmission path; and an average access delay period between the upstream device and the target cache module is an arithmetic mean of a proximal region access delay and a distal region access delay. The method effectively reduces the average access delay period of the target cache module by physical distance partitioning and differential pipeline path design.

This invention discloses a drilling path planning method, storage medium, and device for cochlear implantation, belonging to the field of path planning technology in cochlear implantation. This invention solves the problems of low efficiency and easy introduction of path deviation in existing path planning methods. The invention preprocesses preoperative three-dimensional CT images and integrates anatomical structural models to extract three-dimensional spatial geometric information of key tissues within the temporal bone, including the facial nerve and cochlea. Subsequently, combined with prior medical knowledge, the drilling path for the cochlear implant electrode is automatically planned in three-dimensional space to minimize damage to key nerves and ensure surgical safety. The path planning method of this invention is applicable to cochlear implantation surgery assisted by medical robots, significantly reducing reliance on the doctor's subjective judgment, improving the individualized adaptability and efficiency of path design, and effectively avoiding path deviation caused by human subjective judgment. This invention can be applied to path planning in cochlear implantation surgery.

The invention discloses a method for designing a gluing path of a complex bonding surface, which comprises the following steps: initializing the pose of an end effector of a mechanical arm, and determining the spatial layout and type of a curved surface to be covered; the curved surface to be covered is mapped to a uv space, and a uniform uv grid is generated; establishing a task constraint model of an end effector of the mechanical arm; the coverage feasibility of the curved surface to be covered is checked through a mechanical arm joint space searching method; generating a covering path pattern on the uniform uv grid of the to-be-covered curved surface passing the feasibility test to obtain a gluing path point sequence of the to-be-covered curved surface; and based on the gluing path point sequence of the to-be-covered curved surface, an initial gluing covering path of an end effector of the mechanical arm is obtained, and finally singularity detection and singularity avoidance are carried out to obtain a final non-singularity gluing covering path. The method is suitable for automatic generation of the uniform coverage path of the complex three-dimensional free-form surface under task constraint, and the safety and reliability of surface gluing operation of the spacecraft component are remarkably improved.

This application discloses a method for training an encoding / decoding model, a computer device, and a computer storage medium. By introducing a dual-path design that separates the main encoding structure and the detail encoding structure, differentiated encoding of the main information and detail information in the audio signal is achieved. The main encoding structure focuses on processing the raw spectral data that carries the basic outline and core content of the audio, ensuring accurate capture and efficient encoding of the main information; the detail encoding structure is specifically designed for fine processing of the detail spectral data extracted from the raw spectrum, focusing on preserving subtle features that are crucial to the auditory experience, such as high-frequency overtones, transient changes, and weak environmental noise. This clearly defined structural design avoids the problems of insufficient learning or overfitting that easily occur when a single encoding path processes a large amount of main and detail information simultaneously at high bit rates, thereby effectively reducing the "illusion" phenomenon caused by insufficient learning of detail information or scarcity of training data.

The present application relates to a kind of medical image segmentation method and system based on double encoder and double path fusion decoder.The method is respectively inputted two encoders to preprocessed image: local structure enhancement and feature extractor utilize SCEM to promote boundary contrast, organ structure characteristics are combined with MSDE capturer;Global anatomical context structure extractor is based on PVT modeling global semantic dependence.Subsequently, using HCAM to modulate the attention of double encoder characteristics and depth fusion, then realize cross-level global interaction by MDFR-HFM.Finally, DPF-Decoder is generated through the dual-path design of the two paths of gradual fusion and global direct fusion, local details and global semantics are considered, and the final segmentation map is generated.The present application significantly improves the accuracy, robustness and generalization ability of abdominal multi-organ segmentation.