FIELD: physics, processing of images.
SUBSTANCE: invention refers to the sphere of liquid-crystal display (LCD) control immediately dealing with a method of automatic estimate of LCD data display correctness. Attainment of the result desired is enabled by import of LCD images into a PC memory with the help of a web-camera which is done within the framework of testing activities. The PC generates a neuron network specially trained for each LCD type. Prior to commencement of testing proper the LCD image to have been transmitted undergoes digital filtering. After that the polychrome image is converted into a monochrome (black-and-white) one under an adaptive algorithm with the connected regions of the resultant monochrome image duly identified and analysed. Based on the connected regions analysis one specifies the set of morphometric characteristics used for classification of regions as per configuration, position and orientation. Then, relying on the morphometric characteristics specified, one carries out programme-controlled estimate of the LCD dimensions and position relative to the plane point taken to be the coordinate system zero which information is subsequently taken into consideration when the image imported into the PC memory is processed. Comparison is carried out automatically with the help of the neuron network stored in the PC memory and specially trained to identify the areas of the LCD work surface image that contain the pre-specified connected region morphometric characteristics (with every LCD type). Digital filtering of the image transmitted followed by conversion of the polychrome image into a monochrome (black-and-white) one under an adaptive algorithm as well as the image connected regions identification and analysis and morphometric characteristics specification provides for elimination of LCD lighting irregularity interference with the testing and eventually enables programme-controlled estimate of the LCD dimensions and position relative to the plane point taken to be the coordinate system zero regardless of the LCD occasional turn or displacement which, combined with usage of a trainable neuron network for comparison of the tested and the reference images in terms of morphometric characteristics ensures possibility of high-quality automatic testing of various LCD types without modification of the existing programme code which ultimately results in enhanced testing quality and extended functional capabilities of the technique.
EFFECT: extension of the technique functional capabilities combined with testing quality enhancement and control simplification and cheapening.
5 dwg
Title | Year | Author | Number |
---|---|---|---|
METHOD FOR RECOGNITION AND CLASSIFICATION OF OBJECT SHAPE IN LABYRINTH DOMAIN STRUCTURES | 2012 |
|
RU2522869C2 |
DIAGNOSTIC TECHNIQUE FOR MALIGNANT PLEURAL EFFUSION | 2018 |
|
RU2698909C1 |
SYSTEM OF AUTOMATED VIRTUAL DESIGNING OF CLOTHES FOR USERS AND METHOD OF ITS IMPLEMENTATION | 2019 |
|
RU2718362C1 |
METHOD FOR DETECTING ONCOLOGICAL DISEASES IN PELVIC ORGANS AND SYSTEM FOR IMPLEMENTING METHOD | 2023 |
|
RU2814790C1 |
METHOD OF TEACHING AND MASTERING INSTRUCTION MATERIAL | 2008 |
|
RU2413298C2 |
DEVICE FOR MEASURING DYNAMIC CONTACT ANGLE IN A CHANNEL | 2022 |
|
RU2794420C1 |
SYSTEM FOR DIAGNOSING MALIGNANT AND BENIGN SKIN NEW GROWTHS | 2022 |
|
RU2814539C1 |
SOFTWARE AND HARDWARE COMPLEX DESIGNED FOR PROCESSING AEROSPACE IMAGE OF TERRAIN FOR PURPOSE OF DETECTION, LOCALIZATION AND CLASSIFICATION BY TYPE OF AVIATION AND LAND EQUIPMENT | 2021 |
|
RU2811357C2 |
EXTRACTION OF MULTIPLE DOCUMENTS FROM A SINGLE IMAGE | 2020 |
|
RU2764705C1 |
METHOD OF GENERATING DATA FILE FOR ECG ANALYSIS | 2023 |
|
RU2822304C1 |
Authors
Dates
2009-02-10—Published
2007-06-04—Filed