To enhance the measurement accuracy and robustness of 3D reconstruction systems for obtaining precise shape and position information of object surfaces, including resistance to texture interference, we propose an anti-interference structured light measurement system based on scene surface. This system addresses limitations in stripe-encoded structured light 3D measurement, such as abrupt surface reflectance changes and camera defocusing, while investigating methods for phase retrieval and surface reflectance-based phase compensation. By adjusting the projection modulation intensity of the projector, we control the intensity of the reflected light on the measurement surface, reducing the sensitivity of the camera defocusing blur coefficient to texture variations. We propose an adaptive modulation intensity adjustment method to minimize phase errors between neighboring pixels, enabling high-precision phase retrieval. Experimental results demonstrate that in simulated experiments, the root mean square errors before and after anti-interference processing are 0.2124 rad and 0.0371 rad, respectively. In real-world scenarios, the errors of different measured surfaces decrease by approximately 60%. Comparative experiments validate the effectiveness and feasibility of the proposed method, indicating a significant improvement in measurement accuracy. The method enhances the performance of the coded structured light in applications such as industrial measurement and precise sorting, exhibiting high robustness, accuracy, and anti-interference capabilities.
The harvesting robot for Agaricus bisporus can significantly improve production efficiency. However, due to the limited growing space and complex environment of Agaricus bisporus, they usually grow into clusters with many adhesions and occlusions. This leads to significant stitching and target recognition challenges for Agaricus bisporus cultivation scenes. The high-precision scene stitching and recognition method, based on depth sensing, was proposed and evaluated for the harvest of Agaricus bisporus. The rapid depth map stitching algorithm based on disparity correction was proposed because the complete dataset cannot be obtained from a single collection due to the limited scene space. The strategy of hierarchical recognition of markers, called the "hierarchical watershed" algorithm, based on depth maps, is proposed to overcome the challenge of cluster occlusion in the mushroom harvest. The complete solution is applied to a robot platform that integrates scene stitching, recognition, positioning, and the target harvest. The platform is equipped with three robotic arms to increase harvest speed. The terminal position of the mechanical arm is equipped with suction cup grasping to ensure the quality of the harvest. The results showed that the overall stitching error and center coordinate positioning error were less than 2 𝑚𝑚 in the 200 𝑚𝑚 × 400 𝑚𝑚, and the success rate of picking was 95.82%.
Binocular optical tracking systems are an important component in the field of 3D measurement and reconstruction, and the calibration of objects with reflective marker points attached, hereafter referred to as marker object, is a key factor in the overall measurement accuracy. In response to the challenge of using more sophisticated and expensive instruments such as CMM for the calibration of traditional marker object, this paper proposes a short baseline, high-precision, fast, and low-cost marker object calibration method based on transformation constraints. A binocular calibration system with a short baseline and small field of view is designed to improve the initial spatial resolution accuracy. The multi-angle projection of marker points on the marker object under known fixed transformation constraints is collected through precision servo rotary stage control, and a global error optimization method based on Newton's iterative method is proposed to reduce the estimation error of the initial marker points. The marker object calibration system built in this paper enables a spatial position accuracy resolution of 0.15mm between marker points on the marker object, realizing the need for low-cost, fast and high-precision calibration, and achieving high-precision tracking of binocular optical tracking systems.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.