Abstract: A laparoscopic image smoke removal method based on a generative adversarial network, and belongs to the technical field of computer vision. The method includes: processing a laparoscopic image sample to be processed using a smoke mask segmentation network to acquire a smoke mask image; inputting the laparoscopic image sample to be processed and the smoke mask image into a smoke removal network, and extracting features of the laparoscopic image sample to be processed using a multi-level smoke feature extractor to acquire a light smoke feature vector and a heavy smoke feature vector; and acquiring, according to the light smoke feature vector, the heavy smoke feature vector and the smoke mask image, a smoke-free laparoscopic image by filtering out smoke information and maintaining a laparoscopic image by using a mask shielding effect. The method has the technical effects of robustness and ability of being embedded into a laparoscopic device for use.

Abstract: A laparoscopic image smoke removal method based on a generative adversarial network, and belongs to the technical field of computer vision. The method includes: processing a laparoscopic image sample to be processed using a smoke mask segmentation network to acquire a smoke mask image; inputting the laparoscopic image sample to be processed and the smoke mask image into a smoke removal network, and extracting features of the laparoscopic image sample to be processed using a multi-level smoke feature extractor to acquire a light smoke feature vector and a heavy smoke feature vector; and acquiring, according to the light smoke feature vector, the heavy smoke feature vector and the smoke mask image, a smoke-free laparoscopic image by filtering out smoke information and maintaining a laparoscopic image by using a mask shielding effect. The method has the technical effects of robustness and ability of being embedded into a laparoscopic device for use.

Abstract: A method and system for segmenting three-dimensional images of pancreases and tumors, including: acquiring 3D images of the pancreas and the tumor and preprocessing same with a soft tissue window to control the intensity value of the image within a set range; cropping all images into block-shaped regions of a set size and feeding same into a trained convolutional neural network model, and when training a network, dynamically adjusting the learning weight between the pancreas and the tumor under the guidance of temperature to learn the features of the pancreas and the tumor; and after performing data preprocessing on the 3D images of the pancreas and labels, using the trained network model for online testing and evaluating, and outputting the segmentation result.

Abstract: The present invention provides a method for fabricating KTP nonlinear racetrack micro-ring resonator, composed of six steps: KTP wafer processing, ion implantation, electron beam exposure, subsequent processing, reactive ion etching and final processing. A thin-film waveguide structure similar to the on-insulator lithium niobate thin-film can be achieved through only one process of ion implantation, which enables significantly simplified procedure, shortened time, and reduced cost. Meanwhile, the KTP micro-ring resonator produced according to the present invention has an optical damage threshold several times higher than the existing lithium niobate micro-ring resonator. It can output nonlinear frequency converted light to the power of milliwatts, and suitable for the case where both the input and output optical signals are pulsed lasers.

Abstract: Disclosed are a method and a system for comprehensive evaluation of organic compound and heavy metal pollution in water based on fish electro-cardio, and fish electro-cardio signals are acquired by a real-time and miniaturized fish electro-cardio acquisition system which includes a real-time and miniaturized fish electro-cardio acquisition device, then a change of the electro-cardio index in a QT interval is obtained for assessing the corresponding organic compound in water to be tested, and a change of the electro-cardio index in a QRS interval is obtained for assessing the corresponding heavy metal in water to be tested. Based on fish electro-cardio acquired continuously on-line in real-time while keeping fish swims in a normal state and the water quality parameters acquired by various water quality sensors, water quality is online analyzed and water sudden pollution is online monitored and assessed.

Abstract: A breaststroke leg strength training device, including a lower panel, and a prostrate platform connected with the panel through an elastic band and located below one side of the panel; a slit consistent with a breaststroke leg movement trajectory is formed in the panel, and an elastic band attachment column is respectively arranged at positions on both sides of the slit and on the panel; two foot pedal middle shafts are arranged in the slit and extend to the lower side of the panel; and the elastic band is connected with the attachment columns through a fixed beam fixed to the platform and the two foot pedal middle shafts. The device has a smart structural design, a trainer can comprehensively train all the muscle groups involved in the breaststroke leg actions in a continuous and efficient manner, and the breaststroke leg actions of the trainer can be specified and corrected.

Abstract: A breaststroke leg strength training device, including a lower panel, and a prostrate platform connected with the panel through an elastic band and located below one side of the panel; a slit consistent with a breaststroke leg movement trajectory is formed in the panel, and an elastic band attachment column is respectively arranged at positions on both sides of the slit and on the panel; two foot pedal middle shafts are arranged in the slit and extend to the lower side of the panel; and the elastic band is connected with the attachment columns through a fixed beam fixed to the platform and the two foot pedal middle shafts. The device has a smart structural design, a trainer can comprehensively train all the muscle groups involved in the breaststroke leg actions in a continuous and efficient manner, and the breaststroke leg actions of the trainer can be specified and corrected.