10 results about "Low vision" patented technology

The application relates to a zero-character full-symbol medication guidance system based on a fool-proof design, and belongs to the technical field of medical safety. The system comprises seven pure graphical symbol modules of medication frequency, medication time, medication dose, medication route, contraindication warning, adverse reaction treatment and missed medication treatment. The modules are visually distinguished by unified color, shape and line type, and are combined into a complete information chain in a fixed order of frequency, time, dose, route, contraindication, adverse reaction and missed medication, and are printed or displayed on a medicine box, a medication guidance sheet or a prescription carrier. The application migrates the picture priority and fool-proof and error-proof operation national standard system in the construction industry to a medical medication scene, fundamentally eliminates medication misreading risks caused by ambiguous characters, character ambiguity and cultural level differences, realizes the instinctive recognition and accurate execution of the whole population such as illiterates, the elderly with low vision and cognitive impairment, and has the remarkable technical effects of zero learning cost, full-process coverage and low cost and easy popularization.

1. Name of the product in this design: Vision imaging device. 2. Purpose of this design: This design is intended to be used as a visual magnification device for people with low vision when reading, and the screen can be adjusted to landscape or portrait mode as needed. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model.

A workpiece front and back surface auxiliary material and flatness inspection device includes a conveying device and a transfer device, and further includes: a fixture device, which is disposed beside the conveying device, including a positioning fixture and a first linear module that drives the positioning fixture to reciprocate; the positioning fixture includes a fixture plate and a plurality of positioning components disposed on the fixture plate and defining a clamping area, and a window is also provided on the fixture plate in the clamping area for exposing the lower surface of the workpiece to be inspected; and an inspection device, which is disposed on the moving path of the positioning fixture, and includes an upper vision camera assembly and a lower vision camera assembly symmetrically arranged in the vertical direction. This technical solution can greatly improve the efficiency of inspecting double-sided workpieces, while also reducing the degree of manual intervention and lowering the labor intensity of workers.

The application provides a classroom video character segmentation method based on deep learning, and belongs to the technical field of artificial intelligence, and comprises the following steps: acquiring classroom teaching panoramic video stream data; performing student-level mask segmentation processing on the classroom teaching panoramic video stream data based on a preset multi-scale student segmentation network model, to obtain a student-level mask segmentation prediction result output by the multi-scale student segmentation network model; wherein the multi-scale student segmentation network model is obtained based on deep learning model training of a multi-scale student segmentation network based on spatial semantic guidance on classroom video training data. The technical scheme can enable low-vision users to accurately segment student characters in a classroom video.

This invention discloses an artificial intelligence-based low-vision visual assistance system, belonging to the field of artificial intelligence technology. The system includes a visual perception modeling module, an environmental data acquisition module, a multimodal feature extraction module, a blind spot risk assessment module, a perception delay analysis module, a multimodal output optimization module, and an auxiliary signal output module. By acquiring individual user vision parameters, fusing video and audio data to extract multimodal features, a perception accessibility model is constructed and a blind spot risk map is generated. Based on target trajectory prediction and environmental noise levels, the user's perception delay is dynamically calculated. A machine learning algorithm is used to adaptively select the optimal prompting method, and auxiliary signals are output on the user's device in the form of voice, vibration, or image. This invention can achieve early identification and prompting of potential dynamic risk targets within the visual blind spot, possessing high intelligence, personalization, and practicality, significantly superior to existing visual assistance solutions.

Exemplary light control devices and methods provide a regional variation of visual information and sampling (“V-VIS”) of an ocular field of view that improves or stabilizes vision, ameliorates a visual symptom, reduces the rate of vision loss, or reduces the progression of an ophthalmic or neurologic condition, disease, injury or disorder. The V-VIS devices and methods generate a moving aperture effect anterior to a retina that samples and delivers to the retina environmental light from an ocular field of view at a sampling rate between 50 hertz and 50 kilohertz. Certain of these V-VIS devices and methods may be combined with augmented or virtual reality, vision measurement, vision monitoring, or other therapies including, but not limited to, pharmacological, gene, retinal replacement and stem cell therapies.

This invention discloses a circuit board conveying and barcode scanning mechanism, relating to the technical field of barcode conveying equipment. The invention includes two sets of conveying tracks arranged in parallel. These tracks support the lower edge of the circuit board, allowing it to travel in a straight line. The area between the two tracks is a shooting area. An upper vision component and a lower vision component are correspondingly arranged above and below this shooting area. These components are used to capture images of the upper and lower surfaces of the circuit board to obtain QR codes. This invention employs fixedly installed upper / lower cameras with a 400x400mm field of view, covering circuit boards within a 50mm-400mm range. It eliminates the need for moving the camera via a movable track and avoids exhaustive searching for barcodes. Simultaneously, a surface light source ensures uniform brightness across the entire field of view, resulting in high barcode recognition accuracy. This reduces the cost of the moving track equipment and avoids the problems of circuit wear and collision damage caused by long-term reciprocating motion.

The invention relates to the field of medicine, namely to ophthalmology, and can be used in the combined treatment of active retinopathy of prematurity type 1 in zone 1, in children with a gestational age of less than 28 weeks. The technical objective is to develop an effective method for the combined treatment of active retinopathy of prematurity type 1 in zone 1, in children with a gestational age of less than 28 weeks, by performing simultaneous transpupillary laser coagulation of the retina only in the nasal quadrants and introducing an angiogenesis inhibitor into the vitreous cavity. The proposed method solves this problem, involving simultaneous transpupillary laser photocoagulation of avascular zones of the retina in the nasal quadrants only and the injection of an angiogenesis inhibitor into the vitreous cavity. This method prevents the development of tractional retinal detachment, which causes blindness and low vision in children, reactivation of active ROP, and postoperative uveitis, which require additional surgical treatment, in 100% of cases.