With the development of technology, the camera market has introduced binocular camera. So what exactly is binocular camera? What can binocular camera do and what convenience can binocular cameras bring to our lives?
The binocular camera uses the principle of bionics to obtain synchronous exposure images through the calibrated dual cameras, and then calculates the third-dimensional depth information of the acquired 2-dimensional image pixels. In order to monitor different environmental situations, a new adaptive recognition algorithm based on binocular computer vision is proposed. The algorithm first uses the depth information of the pixels to identify and judge the scene, and then uses the statistical method to model the scene, and overcomes the illumination gradient through time filtering, and overcomes the illumination mutation through the characteristics of the depth algorithm. Compared with a single camera, the video surveillance principle system realized by this algorithm can be applied to more occasions, and the alarm level is set by using the depth information to reduce the false detection rate.
What can binocular camera be used for ?
Applications related to distance
The human eye can perceive the distance of an object due to the difference between the images presented by the two eyes to the same object, also known as "parallax". The farther the object is, the smaller the parallax; conversely, the greater the parallax. The magnitude of parallax corresponds to the distance between the object and the eye, which is why 3D movies can make people perceive three-dimensional layers.
It can be seen from the figure that the person is on the left side of the tree in the right image, and the person is on the right side of the tree in the left image, because the binocular angles are different. By comparing the two images, we can know that the parallax is small when the human eye observes the tree, and the parallax is large when observing the person, because the distance between the tree is far and the distance between the person is short. This is the principle of binocular triangulation. The binocular system's perception of the distance to the target object is an absolute measurement, not an estimate.
If the FOV of the two cameras is different, a large FOV and a small FOV, and then through the algorithm to achieve the effect between the two optical lenses, optical zoom can be easily achieved.
The performance of the binocular camera to the ISP is relatively high, the main function of ISP is to post-process the signal output by the front-end image sensor, it includes linear correction, noise removal, dead pixel removal, interpolation, white balance, automatic exposure control, etc., Relying on ISP can restore scene details better under different optical conditions, ISP technology determines the imaging quality of the camera to a large extent.
A good camera algorithm needs to be matched with a good ISP, ISP and algorithm complement each other and are indispensable, the dual camera requires more ISP performance. In order to enable the signals of the left and right cameras to be processed at the same time, a single ISP has been unable to meet the needs of dual cameras, this requires dual ISP to achieve this function. Taking dark light enhancement as an example, the color/black and white images enter their respective ISP channels and calibration channels, and then match the two pictures (for example, extract the same part of the two pictures and remove the part captured by only one camera), then the related pictures are processed through occlusion, detection, compensation and other algorithms, the two images are merged finally to achieve color enhancement.
With the evolution of the dual-camera algorithm, the video surveillance system with dual-camera will also play a greater role and be applied to more scenarios.
