Indoor and outdoor depth imaging of leaves with time-of-flight and stereo vision sensors: Analysis and comparison

Abstract

In this article we analyze the response of Time of Flight cameras (active sensors) for close range imaging under three different illumination conditions and compare the results with stereo vision (passive) sensors. Time of Flight sensors are sensitive to ambient light and have low resolution but deliver high frame rate accurate depth data under suitable conditions. We introduce some metrics for performance evaluation over a small region of interest. Based on these metrics, we analyze and compare depth imaging of leaf under indoor (room) and outdoor (shadow and sunlight) conditions by varying exposures of the sensors. Performance of three different time of flight cameras (PMD CamBoard, PMD CamCube and SwissRanger SR4000) is compared against selected stereo-correspondence algorithms (local correlation and graph cuts). PMD CamCube has better cancellation of sunlight, followed by CamBoard, while SwissRanger SR4000 performs poorly under sunlight. stereo vision is more robust to ambient illumination and provides high resolution depth data but it is constrained by texture of the object along with computational efficiency. Graph cut based stereo correspondence algorithm can better retrieve the shape of the leaves but is computationally much more expensive as compared to local correlation. Finally, we propose a method to increase the dynamic range of the ToF cameras for a scene involving both shadow and sunlight exposures at the same time using camera flags (PMD) or confidence matrix (SwissRanger).

ToF cameras for active vision in robotics

“…Finally, we have tested ToF cameras in outdoor scenarios with sunlight [4]. An algorithm has been proposed to select the best integration time depending on the sun conditions, as well as a suitable strategy to combine two frames to obtain depth images even when a plant is partially illuminated with direct sunlight and partially in shadow, as it is common in greenhouses.…”

“…The main difficulties for most sensors outdoors are caused by sunlight effects. However, it has been demonstrated that ToF cameras can be configured to deliver correct enough images in sunlight [4]. Figure 6 shows the robot facing three different outdoor obstacles: a wall (Fig.…”

Detecting creeping thistle in sugar beet fields using vegetation indices

“…1 show that bandwidth of the filter is narrower for blue and red channels than (Kazmi et al, 2014) using the same camera. It shows the effect of exposure by controlling Shutter Time (ST: the duration of time for which the camera receives the incoming light) on the appearance of a leaf.…”

“…This can reduce the complexity of the system by sparing the computationally expensive task of resolution of plant overlap and leaf segmentation which may require 3D imaging Alenya et al, 2011) and 3D sensing has its own set of challenges, especially when it comes to outdoor scenarios (Kazmi et al, 2014).…”

Plant Localization and Discrimination using 2D+3D Computer Vision for Robotic Intra-row Weed Control

“…There are many research applications using TOF cameras as sensing devices in agriculture, such as phenotyping (Alenyà, Dellen, Foix, & Torras, 2012), and plant spacing (Jin & Tang, 2009). In study by Kazmi, Foix, Alenyà, & Andersen, (2014), the authors summarized the advantages along with drawbacks of TOF cameras. The advantages are: they deliver high frame rates as well as accurate depth data under suitable conditions.…”