The Way Of Interaction Of Smart Glasses

It mainly introduces three interactive methods widely used in smart glasses, namely voice control, gesture recognition and eye tracking.

1. Voice control

In people's daily communication, speaking is the most commonly used method. Introducing voice interaction into the wearable field, people will be able to enjoy a more natural and relaxing interactive experience. Voice control means that the computing device can understand what a person says, and it can also execute corresponding instructions according to the content of the person's speech. For smart glasses that are small in size and worn on the body, voice control is an effective way of interaction.

Principle of Voice Control

The core part of voice control is voice recognition technology. Bone conduction technology can complete the efficient recognition and transmission of voice, and many smart glasses adopt this technology. Take Buhel's Sound Glass as an example. Sound Glass is equipped with an indirect bone conduction sensor. Each of his temples has a sound transducer. The sound produced by the transducer can be transmitted to the inner ear through the bones on the side of the user's head, so that the user can hear the sound. Although voice control is important in smart glasses However, the voice control has encountered many difficulties.

The flaws of voice control

First of all, there are many interfering factors in the extraction of speech signals, such as differences in vocalization between individuals and changes in their own intonation, differences in the way people speak in different regions and cultural backgrounds, and the interference of environmental noise on speech signals, etc. These factors will adversely affect the extraction of speech signals. Secondly, the efficiency and speed of voice recognition need to be improved. These two points directly affect the application value of voice control in smart glasses, and are an important measure of application value. In addition, users have high expectations for voice control, but the actual situation is that voice control cannot meet the needs of users. For example, when a user initiates a voice control command using Google Glass, the user must strictly follow the standard method provided by Google Glass. When the user wants to make a call, he must say "ok glass, make a call to...", and the more accustomed way "ok glass, call." is completely invalid.

2. Gesture recognition

Using gestures as input to complete the interactive function of smart glasses has the advantage of using a non-contact method. Gesture recognition technology can be divided into three categories from simple and rough to complex and fine: two-dimensional hand recognition, two-dimensional gesture recognition, and three-dimensional gesture recognition. The difference between three-dimensional gesture recognition and two-dimensional gesture recognition is that the input information of three-dimensional gesture recognition also contains depth information. Smart glasses adopting three-dimensional gesture recognition can realize more and more complex interaction methods.

Gesture recognition principle and sensor

Three-dimensional gesture recognition needs depth information to be able to recognize various gestures, hand shapes and actions. To obtain depth information, special hardware is needed, and with the recognition algorithm, three-dimensional gesture recognition can be realized. Next, introduce several special sensors for gesture recognition: TMG399, which is a non-contact optical IR gesture recognition sensor, equipped with a four-in-one sensor module for gesture recognition, ambient light detection, proximity perception and color perception; MGC3130, microchip The 3D gesture recognition chip launched by Science and Technology, under the action of its electric field, can sense gestures without contact, and can determine the coordinate position with a high precision of 150dpi within a distance of 15cm; MYO, a product of the start-up company Thalmic Labs, is a wearable The armband on the arm; 16Lab, this is a smart ring for gesture control, built-in inertial sensor module, processor and low energy Bluetooth module.

Gesture recognition defect

However, gesture recognition has also exposed some shortcomings in the process of being applied to smart glasses. First of all, the accuracy of gesture recognition is low, and the positioning is not accurate enough. Because everyone's hand structure is different, it is difficult to achieve precise positioning by capturing hand movements. Secondly, the key to gesture recognition is the extraction of finger features. It must be able to accurately distinguish the features of the target in a complex background, but it is still difficult to overcome the situation where the gesture is occluded or the removal of redundant information. problem.

3. Eye tracking

Eye tracking is the process of measuring the gaze point of the eye or the movement state of the glasses relative to the head. Google Glass can perceive the user's emotions through eye-tracking technology to determine the user's response to the ad that is watching.

Principles of Eye Tracking

The eye tracking measurement technology used in smart glasses is mainly based on image and video measurement methods. This method includes a variety of techniques for measuring distinguishable eye movement characteristics, such as the heterochromatic edges of the sclera and iris, and the light intensity reflected by the cornea. And the appearance and shape of the pupil. The method based on image, combined with pupil shape change and corneal reflection is widely used in measuring the focus of the user's line of sight.

Eye tracking defects

Although the eyes are the widest and fastest way to receive information in the body, eye tracking is far from the humanized way of interaction. Due to the inherent blinking and jitter of the eyes, a lot of interference signals will be generated, which may cause data interruption, which will greatly increase the difficulty of extracting accurate data from eye movement information.