Mark R. Shortis

A comparison of underwater visual distance estimates made by scuba divers and a stereo-video system: implications for underwater visual census of reef fish abundance

Underwater visual census of reef fish by scuba divers is a widely used and useful technique for assessing the composition and abundance of reef fish assemblages, but suffers from several biases and errors. We compare the accuracy of underwater visual estimates of distance made by novice and experienced scientific divers and an under- water stereo-video system. We demonstrate the potential implications that distance errors may have on underwater visual census assessments of reef fish abundance. We also investigate how the accuracy and precision of scuba diver length estimates of fish is affected as distance increases. Distance was underestimated by both experienced (mean relative error =− 11.7%, s.d. = 21.4%) and novice scientific divers (mean relative error =− 5.0%, s.d. = 17.9%). For experienced scientific divers this error may potentially result in an 82% underestimate or 194% overestimate of the actual area censused, which will affect estimates of fish density. The stereo-video system also underestimated distance but to a much lesser degree (mean relative error =− 0.9%, s.d. = 2.6%) and with less variability than the divers. There was no correlation between the relative error of length estimates and the distance of the fish away from the observer.

Photogrammetry: A Powerful Tool for Geometric Analysis of Solar Concentrators and Their Components

Digital close range photogrammetry has proven to be a precise and efficient measurement technique for the assessment of shape accuracies of solar concentrators and their components. The combination of high quality mega-pixel digital still cameras, appropriate software and calibrated reference scales in general is sufficient to provide coordinate measurements with precisions of 1:50,000 or better. The extreme flexibility of photogrammetry to provide high accuracy 3-D coordinate measurements over almost any scale makes it particularly appropriate for the measurement of solar concentrator systems. It can also provide information for the analysis of curved shapes and surfaces, which can be very difficult to achieve with conventional measurement instruments. The paper gives an overview of quality indicators for photogrammetric networks, which have to be considered during the data evaluation to augment the measurement precision. A selection of measurements done on whole solar concentrators and their components are presented. The potential of photogrammetry is demonstrated by presenting measured effects arising from thermal expansion and gravitational forces on selected components. The measured surface data can be used to calculate slope errors and undertake raytrace studies to compute intercept factors and assess concentrator qualities.

A comparison of the accuracy and precision of measurements from single and stereo-video systems

Underwater tests using plastic silhouettes of fish were used to compare the accuracy and precision of measurements made with a single video camera system to those made from two stereo-video systems (one using digital camcorders, the other using Hi8 camcorders). Test measurements made across a variety of ranges and angles of silhouette orientation in the fields of view showed the length estimates from both the digital and Hi8 stereo-video systems were substantially more accurate and precise than those obtained by the single video camera system, and had the great advantage that the position (range and bearing) and orientation of a fish target could be measured directly. Measurements made with stereo-video were much less restricted by range and subject orientation than those made with single video. The data resulting from these trials are used to propose a set of guidelines to optimize the accuracy and precision of underwater measurements of fish length using single and stereo-video systems.

Photogrammetry Methodology Development for Gossamer Spacecraft Structures

Photogrammetry--the science of calculating 3D object coordinates from images--is a flexible and robust approach for measuring the static and dynamic characteristics of future ultra-lightweight and inflatable space structures (a.k.a., Gossamer structures), such as large membrane reflectors, solar sails, and thin-film solar arrays. Shape and dynamic measurements are required to validate new structural modeling techniques and corresponding analytical models for these unconventional systems. This paper summarizes experiences at NASA Langley Research Center over the past three years to develop or adapt photogrammetry methods for the specific problem of measuring Gossamer space structures. Turnkey industrial photogrammetry systems were not considered a cost-effective choice for this basic research effort because of their high purchase and maintenance costs. Instead, this research uses mainly off-the-shelf digitalcamera and software technologies that are affordable to most organizations and provide acceptable accuracy.

Influence of range, angle of view, image resolution and image compression on underwater stereo-video measurements: High-definition and broadcast-resolution video cameras compared

An investigation of how increasing the distance and angle to objects of interest affected the measurement accuracy and precision achievable with high-definition and medium-resolution PAL stereo-video systems was conducted. A test was also conducted to determine whether varying the compression of the imagery influenced measurement accuracy and precision. Measurements of five different lengths of PVC pipe (ranging from 51.5 to 3,001 mm) that represented the lengths of reef fishes routinely sampled with stereo-video were made at 1 m intervals out to the maximum visibility (9 m range) over three different angles (90o, 80o and 70o). High-definition stereo-video imagery was compressed at three different bit-rates. The results show that higher definition stereo-video imagery allows objects to be measured more accurately and precisely over greater ranges. When both ends of a target can be clearly seen in high-definition stereo-video imagery, the associated error is approximately 1% of the total length of the object. There was no deleterious effect on accuracy or precision from increasing the angle of view. Lower compression did not result in more accurate and precise length estimates. The configuration of a stereo-video system needs to match the task of a particular survey, as changes in the base separation and angle of convergence will affect the accuracy and precision of measurements. With full high-definition systems, smaller lengths (<50 mm) of PVC could not be accurately measured at distances greater than 5 m whereas longer lengths (500-3,001 mm) could be measured with acceptable accuracy and precision at 9 m.

A review of techniques for the identification and measurement of fish in underwater stereo-video image sequences

Underwater stereo-video measurement systems are used widely for counting and measuring fish in aquaculture, fisheries and conservation management. To determine population counts, spatial or temporal frequencies, and age or weight distributions, snout to fork length measurements are captured from the video sequences, most commonly using a point and click process by a human operator. Current research aims to automate the measurement and counting task in order to improve the efficiency of the process and expand the use of stereo-video systems within marine science. A fully automated process will require the detection and identification of candidates for measurement, followed by the snout to fork length measurement, as well as the counting and tracking of fish. This paper presents a review of the techniques used for the detection, identification, measurement, counting and tracking of fish in underwater stereo-video image sequences, including consideration of the changing body shape. The review will analyse the most commonly used approaches, leading to an evaluation of the techniques most likely to be a general solution to the complete process of detection, identification, measurement, counting and tracking.

Calibration Techniques for Accurate Measurements by Underwater Camera Systems

Calibration of a camera system is essential to ensure that image measurements result in accurate estimates of locations and dimensions within the object space. In the underwater environment, the calibration must implicitly or explicitly model and compensate for the refractive effects of waterproof housings and the water medium. This paper reviews the different approaches to the calibration of underwater camera systems in theoretical and practical terms. The accuracy, reliability, validation and stability of underwater camera system calibration are also discussed. Samples of results from published reports are provided to demonstrate the range of possible accuracies for the measurements produced by underwater camera systems.

Image selection in photogrammetric multi-view stereo methods for metric and complete 3D reconstruction

Multi-View Stereo (MVS) as a low cost technique for precise 3D reconstruction can be a rival for laser scanners if the scale of the model is resolved. A fusion of stereo imaging equipment with photogrammetric bundle adjustment and MVS methods, known as photogrammetric MVS, can generate correctly scaled 3D models without using any known object distances. Although a huge number of stereo images (e.g. 200 high resolution images from a small object) captured of the object contains redundant data that allows detailed and accurate 3D reconstruction, the capture and processing time is increased when a vast amount of high resolution images are employed. Moreover, some parts of the object are often missing due to the lack of coverage of all areas. These problems demand a logical selection of the most suitable stereo camera views from the large image dataset. This paper presents a method for clustering and choosing optimal stereo or optionally single images from a large image dataset. The approach focusses on the two key steps of image clustering and iterative image selection. The method is developed within a software application called Imaging Network Designer (IND) and tested by the 3D recording of a gearbox and three metric reference objects. A comparison is made between IND and CMVS, which is a free package for selecting vantage images. The final 3D models obtained from the IND and CMVS approaches are compared with datasets generated with an MMDx Nikon Laser scanner. Results demonstrate that IND can provide a better image selection for MVS than CMVS in terms of surface coordinate uncertainty and completeness.

Towards fully automatic reliable 3D acquisition

This paper describes a novel system for accurate 3D digitization of complex objects. Its main novelties can be seen in the new approach, which brings together different systems and tools in a unique platform capable of automatically generating an accurate and complete model for an object of interest. This is performed through generating an approximate model of the object, designing a stereo imaging network for the object with this model and capturing the images at the designed postures through exploiting an inverse kinematics method for a non-standard six degree of freedom robot. The images are then used for accurate and dense 3D reconstruction using photogrammetric multi-view stereo method in two modes, including resolving scale with baseline and with control points. The results confirm the feasibility of using Particle Swarm Optimization in solving inverse kinematics for this non-standard robot. The system provides this opportunity to test the effect of incidence angle on imaging network design and shows that the matching algorithms work effectively for incidence angle of 10?. The accuracy of the final point cloud generated with the system was tested in two modes through a comparison with a dataset generated with a close range 3D colour laser scanner. We designed a 6-DOF robot for accurate and dense 3D reconstruction using images.Particle Swarm Optimization was evaluated for inverse kinematic of the robot.A software package, Imaging Network Designer, was tested for this robot.The accuracy of the robot for 3D reconstruction is around 200 µ m .

Automatic fish species classification in underwater videos: Exploiting pre-trained deep neural network models to compensate for limited labelled data

There is a need for automatic systems that can reliably detect, track and classify fish and other marine species in underwater videos without human intervention. Conventional computer vision techniques do not perform well in underwater conditions where the background is complex and the shape and textural features of fish are subtle. Data-driven classification models like neural networks require a huge amount of labelled data, otherwise they tend to over-fit to the training data and fail on unseen test data which is not involved in training. We present a state-of-the-art computer vision method for fine-grained fish species classification based on deep learning techniques. A cross-layer pooling algorithm using a pre-trained Convolutional Neural Network as a generalized feature detector is proposed, thus avoiding the need for a large amount of training data. Classification on test data is performed by a SVM on the features computed through the proposed method, resulting in classification accuracy of 94.3% for fish species from typical underwater video imagery captured off the coast of Western Australia. This research advocates that the development of automated classification systems which can identify fish from underwater video imagery is feasible and a cost-effective alternative to manual identification by humans.