A device for identifying certain pointing information or image-text information

Abstract

The invention discloses a device for identifying certain pointing information or image-text information. The system comprises a photographing module and a monitoring module. the photographing module comprises a camera; when the monitoring module is triggered; the pointing information or image-text information identification equipment shoots downwards through the camera to obtain a book image to shoot and identify the book; the monitoring module comprises an infrared or ultrasonic monitoring unit; the infrared or ultrasonic monitoring unit is an infrared or ultrasonic transmitting and receivinggroup; wherein the triggering of the monitoring module comprises triggering by transmitting and receiving reflected infrared rays or ultrasonic waves through the infrared or ultrasonic monitoring unit, the infrared or ultrasonic monitoring unit comprises a first infrared or ultrasonic monitoring unit, and the monitoring distance of the first infrared or ultrasonic monitoring unit is adjustable. Implementation of the embodiment of the invention is favorable for identifying the pointing point of a hand on a book or identifying which book or page.

Description

technical field [0001] The invention relates to the field of electronic technology, in particular to image-text recognition. Background technique [0002] At present, there is no actual product that relies on the camera to achieve point-and-click reading, especially no actual product that can take a photo to identify which page and position the pointing is on by pointing directly on the book, nor can it be performed by directly pointing the finger on the graphic text Take pictures to identify the actual product of the picture and text, [0003] Some of the current schemes cannot be converted into practical products. These schemes rely entirely on image recognition of continuous photographs. The disadvantage is that, on the one hand, the accuracy of photographic recognition in complex book backgrounds is low, and it is different from other simple backgrounds. Books have different patterns, and some patterns are similar to fingers or sticks and are easily confused. Recognitio...

AI Technical Summary

Problems solved by technology

[0003] Some of the current schemes cannot be converted into practical products. These schemes rely entirely on image recognition of continuous photographs. The disadvantage is that, on the one hand, the accuracy of photographic recognition in complex book backgrounds is low, and it is different from other simple backgrounds. Books have different patterns, and some patterns are similar to fingers or sticks and are easily confused. Recognition is entirely based on images. In practical applications, there are often no pointing but misidentified as pointing, or it takes a long time to reliably confirm in practical applications. Is there any guidance? On the other hand, most of the time, the user does not perform any operations, neither turning the page nor giving guidance. At this time, the continuous photo recognition is invalid and inefficient, and it will also occupy the processor memory of the device and cause other work. It can't be done well. If you want to take continuous and invalid photos while completing other tasks, the hardware requirements are too high. The system is too complicated and the cost is too high, and it is impossible to form a practical product.

[0004] Therefore, there is currently no photo recognition of the actual product that is directly pointed on the book with your finger

Images