Author name cluster

Arto Visala

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers

2 author rows

IROS Conference 2021 Conference Paper

Shipborne sea-ice field mapping using a LiDAR

Andrei Sandru
Arto Visala
Pentti Kujala

The increasing interest for autonomous ships has motivated research in numerous areas. One such area is the safe navigation through ice infested waters, for which a sensor instrumentation and automated process are proposed for near-field, sea-ice 3D scanning and mapping using a ship mounted LiDAR, with attitude compensation from inertial and satellite positioning sensors. Data were collected both at the Aalto Ice Tank laboratory and on board the icebreaker S. A. Agulhas II during its voyage to the Antarctic waters. The implemented process enables automated acquisition of detailed 3D point cloud maps, containing highly valuable information for icy waters going ships currently operated by a human crew and, in the near future, supporting the development of autonomous ships. Compared to other methods using satellite, aerial or underwater data, the proposed method is a more cost-effective and easy to integrate solution into current and future icy waters going ships, thus enabling a higher level of situational awareness.

Details

AAAI Conference 2020 Conference Paper

Optical Flow in Deep Visual Tracking

Mikko Vihlman
Arto Visala

Single-target tracking of generic objects is a difﬁcult task since a trained tracker is given information present only in the ﬁrst frame of a video. In recent years, increasingly many trackers have been based on deep neural networks that learn generic features relevant for tracking. This paper argues that deep architectures are often ﬁt to learn implicit representations of optical ﬂow. Optical ﬂow is intuitively useful for tracking, but most deep trackers must learn it implicitly. This paper is among the ﬁrst to study the role of optical ﬂow in deep visual tracking. The architecture of a typical tracker is modiﬁed to reveal the presence of implicit representations of optical ﬂow and to assess the effect of using the ﬂow information more explicitly. The results show that the considered network learns implicitly an effective representation of optical ﬂow. The implicit representation can be replaced by an explicit ﬂow input without a notable effect on performance. Using the implicit and explicit representations at the same time does not improve tracking accuracy. The explicit ﬂow input could allow constructing lighter networks for tracking.

PDF Details

ICRA Conference 2015 Conference Paper

Detection and species classification of young trees using machine perception for a semi-autonomous forest machine

Mikko Vihlman
Heikki Hyyti
Jouko Kalmari
Arto Visala

An approach to automatically detect and classify young spruce and birch trees in forest environment is presented. The method could be used in autonomous or semi-autonomous forest machines during tending operations. Detection is done by segmenting laser range images formed by a rotating laser scanner. Classification is done with a two-class Naive Bayes classifier based on image texture features. Multiple combinations of 99 features were tested and the best classifier included eight features from the co-occurrence matrix, local binary patterns, statistical geometrical features and Gabor filter. 79% of spruces and birches in the testing material were detected and 74% of these were correctly classified. Results suggest that the approach is suitable but there are still some challenges in each of the processing steps. Iteration between segmentation and classification is needed to increase reliability.

Details

ICRA Conference 2013 Conference Paper

Stereo vision based tree planting spot detection

Teemu Kemppainen
Arto Visala

Forest planting currently is expensive and very labor intensive. In order to automatize forest planting, it is necessary to detect good planting spots automatically. In this paper, a method for automatic detection of planting spots is described. The information for the detection is deduced merely from stereo images. A continuous 3D reconstruction is performed and the planting spots are inferred using a 3D Implicit Shape Model and planting spot score boosting. The results of our method are promising: the precision of the detector is 94. 8 % with a recall rate of 98. 0 %.

Details

ICRA Conference 2007 Conference Paper

Simultaneous Localization and Mapping for Forest Harvesters

Mikko Miettinen
Matti Öhman
Arto Visala
Pekka Forsman

A real-time SLAM (simultaneous localization and mapping) approach to harvester localization and tree map generation in forest environments is presented in this paper. The method combines 2D laser localization and mapping with GPS information to form global tree maps. Building an incremental map while also using it for localization is the only way a mobile robot can navigate in large outdoor environments. Until recently SLAM has only been confined to small-scale, mostly indoor, environments. We try to addresses the issues of scale for practical implementations of SLAM in extensive outdoor environments. Presented algorithms are tested in real outdoor environments using an all-terrain vehicle equipped with the navigation sensors and a DGPS receiver.

Details