5results about How to "Improve post-processing efficiency" patented technology

This utility model provides a sample post-processing system, including a sample entry / exit unit, a transport track, a refrigeration module, a conveying module, and a scheduling module. The conveying module interfaces with the refrigeration module, enabling the storage of samples in the refrigeration module and the transfer of samples from the refrigeration module. The scheduling module coordinates the arrangement of sample containers on the conveying module, allowing samples to be transported back to the production line for re-inspection. Through the cooperation of the conveying module and the scheduling module, the system achieves sample transport, storage in the refrigeration module, and re-inspection of samples transferred from the refrigeration module. This allows sample containers to circulate smoothly between the sample entry / exit unit, the transport track, and the refrigeration module, simplifying the sample post-processing process.

The invention relates to a YOLO v3 post-processing hardware acceleration method based on a look-up table method, which comprises the following steps of: 1) establishing two look-up tables according to two operators of sigmoid and exp functions involved in a decoding calculation process; 2) analyzing the required storage space according to the quantization bits and the network structure; 3) distributing a corresponding BRAM storage block for the lookup table according to the required storage space; 4) calculating the data to be decoded, and putting a calculation result into a lookup table; 5) inputting to-be-decoded data, putting the to-be-decoded data into the cache unit, and arranging the to-be-decoded data; 6) reading data to be decoded, and searching corresponding calculation data from the lookup table; 7) inputting the calculation data into a parallel calculation unit to obtain decoding data; and 8) putting the decoded data back to the initial position in the cache unit. According to the method, the lookup table is established, the to-be-decoded data in YOLO v3 post-processing is subjected to decoding calculation, post-processing calculation is completed under the condition that few storage resources are used, the hardware scale is remarkably reduced, and the post-processing efficiency is improved.

The invention provides aluminum fumarate and a synthesis method thereof. The synthesis method comprises the following steps: mixing organic aluminum, an organic carboxylic acid ligand and an alkaline substance in water according to a molar ratio of 1: (1.8-2.8): (1-8), and carrying out thermal reaction to obtain aluminum fumarate; wherein the organic carboxylic acid ligand is selected from one or a combination of more than two of fumaric acid, maleic acid and isophthalic acid. According to the synthesis method, organic aluminum is adopted, a large amount of heat can be released when the organic aluminum is decomposed in water, so that synthesis of aluminum fumarate is promoted, meanwhile, the decomposition process is beneficial to promotion of deprotonation of fumaric acid, reaction time is shortened, the dosage of alkaline substances is reduced, generated wastewater is closer to be neutral, cost is reduced, and the production scale is convenient to enlarge. The aluminum fumarate is of a staggered flower-shaped spherical structure, and the specific surface area of the aluminum fumarate is remarkably increased.

The invention discloses an automatic shooting system based on a wireless positioning beacon, relates to the technical field of automatic shooting, and aims to solve the problems of high cost, complex deployment, low multi-target tracking efficiency and difficulty in classification of shot contents of an existing automatic shooting system. The system comprises a plurality of beacons and one or more shooting devices, the beacons are attached to a shot object, and distance measurement, communication and motion sensor modules are integrated; the shooting device integrates distance measurement, communication, image recognition, shooting, track setting and a main program module. And the beacon and the shooting device dynamically adjust the distance measurement frequency through non-accurate time synchronization, and realize multi-target tracking, automatic shooting and shooting content classification in combination with the distance change characteristics of the preset motion trail and the image recognition result. The system is low in cost, easy to deploy, efficient in multi-target tracking, accurate in shot content classification and suitable for specific scenes such as ski field racing tracks and running roads.

The application belongs to the technical field of underwater robot visual perception, edge computing and target detection, and discloses an underwater target automatic identification method and system based on an improved YOLO algorithm, wherein the post-processing speed is greatly improved, an INT8 domain pre-confidence screening mechanism filters more than 90% of background candidate boxes, the time consumption ratio of post-processing is reduced from more than 30% of the traditional to less than 10%, and the overall operation speed of the application is significantly improved. The application has good real-time performance, a multi-thread parallel inference pipeline architecture fully utilizes the heterogeneous computing power of RK3588S, hides single-frame inference delay, realizes stable real-time detection frame rate of 15-25 FPS under 1280*720 resolution, and can timely capture fast-moving underwater targets. The application has high integration, realizes deep integration of the identification application and technologies such as mechanical grabbing and autonomous return, forms a full-autonomous operation closed loop of identification-positioning-grabbing, and the autonomous decision rate is more than 90%, thereby greatly reducing the dependence on manual operation.