Log End Research Articles

Log Volume Measurement and Counting Based on Improved Cascade Mask R-CNN and Deep SORT

Logs require multiple verifications to ensure accurate volume and quantity measurements. Log end detection is a crucial step in measuring log volume and counting logs. Currently, this task primarily relies on the Mask R-CNN instance segmentation model. However, the Feature Pyramid Network (FPN) in Mask R-CNN may compromise accuracy due to feature redundancy during multi-scale fusion, particularly with small objects. Moreover, counting logs in a single image is challenging due to their large size and stacking. To address the above issues, we propose an improved log segmentation model based on Cascade Mask R-CNN. This method uses ResNet for multi-scale feature extraction and integrates a hierarchical Convolutional Block Attention Module (CBAM) to refine feature weights and enhance object emphasis. Then, a Region Proposal Network (RPN) is employed to generate log segmentation proposals. Finally, combined with Deep SORT, the model tracks log ends in video streams and counts the number of logs in the stack. Experiments demonstrate the effectiveness of our method, achieving an average precision (AP) of 82.3, APs of 75.3 for small, APm of 70.9 for medium, and APl of 86.2 for large objects. These results represent improvements of 1.8%, 3.7%, 2.6%, and 1.4% over Mask R-CNN, respectively. The detection rate reached 98.6%, with a counting accuracy of 95%. Compared to manually measured volumes, our method shows a low error rate of 4.07%.

A Comparative Literature Review of Machine Learning and Image Processing Techniques Used for Scaling and Grading of Wood Logs

This literature review assesses the efficacy of image-processing techniques and machine-learning models in computer vision for wood log grading and scaling. Four searches were conducted in four scientific databases, yielding a total of 1288 results, which were narrowed down to 33 relevant studies. The studies were categorized according to their goals, including log end grading, log side grading, individual log scaling, log pile scaling, and log segmentation. The studies were compared based on the input used, choice of model, model performance, and level of autonomy. This review found a preference for images over point cloud representations for logs and an increase in camera use over laser scanners. It identified three primary model types: classical image-processing algorithms, deep learning models, and other machine learning models. However, comparing performance across studies proved challenging due to varying goals and metrics. Deep learning models showed better performance in the log pile scaling and log segmentation goal categories. Cameras were found to have become more popular over time compared to laser scanners, possibly due to stereovision cameras taking over for laser scanners for sampling point cloud datasets. Classical image-processing algorithms were consistently used, deep learning models gained prominence in 2018, and other machine learning models were used in studies published between 2010 and 2018.

Assessing Treatments to Mitigate End-Face Cracking in Air-Dried Acacia dealbata Logs

Acacia dealbata Link, known as Mimosa in Portugal, is an invasive hardwood species with potential for construction use, but research is limited. The available stock of small-diameter juvenile wood logs can help reduce this gap, but tangential cracking at log ends challenges fastener connections. This study evaluated different treatments to control and reduce end-face cracking in small wood logs during air drying, an economical and environmentally friendly procedure. The extreme two-thirds of sixteen Mimosa logs were subjected to two treatments: one with longitudinal kerfs 15 mm deep along the length (two and three kerfs) and the other with a hollow in the center up to half the length (16 mm and 30 mm diameters). Over 219 days of air drying and compared with the central part, kerfing treatments significantly reduced outer-wood tangential cracking (p < 0.001), with the three kerfs also reducing crack numbers (p < 0.05) but increasing significantly cracks near the pith (p < 0.01). The 30 mm central hollow significantly reduced central perforation cracking (p < 0.05). Prospectively, the results suggest that a combined treatment approach involving cross-cuts could help reduce and/or control end cracking, thereby improving the suitability of wild Mimosa logs for construction use.

Image generation of log ends and patches of log ends with controlled properties using generative adversarial networks

The appearance of the log cross-section provides important information when assessing the quality of the log, where properties to consider include pith location and density of annual rings. This makes tasks like estimation of pith location and annual ring detection of great interest. However, creating labeled training data for these tasks can be time-consuming and subject to misjudgments. For this reason, we aim to create generated training data with controlled properties of pith location and amount of annual rings. We propose a two-step generator based on generative adversarial networks in which we can completely avoid manual labeling, not only when generating training data but also during training of the generator itself. This opens up the possibility to train the generator on other types of log end data without the need to manually label new training data. The same method is used to create two generated training datasets; one of entire log ends and one of patches of log ends. To evaluate how the generated data compares to real data, we train two deep learning models to perform estimation of pith location and ring counting, respectively. The models are trained separately on real and generated data and evaluated on real data only. The results show that the performance of both estimation of pith location and ring counting can be improved by replacing real training data with larger sets of generated training data.

An accurate handheld device to measure log diameter and volume using machine vision technique

Log diameter and volume measurements are crucial processes in the wood supply chain. Conventionally, the measurement mainly focuses on diameters and lengths. Despite the clear definition of conventional measurement methods, obtaining exact diameters is challenging due to the complexity of the measuring process and variability in the organic nature of logs. In this study, a smart handheld device, mainly composed of a single industrial camera, laser projectors, and a small embedded processor, is developed to capture and process log end images. Additionally, a measuring method is proposed, which consists of instance segmentation, contour fitting, and pose estimation, to automatically measure the diameters and estimate the volumes of cut-to-length (CTL) logs by using formulae. Specifically, instance segmentation is used to identify the salient object as a contour in the image space. Contour fitting is then applied to fit the contour into an ellipse, where the major and minor axes represent the long and short diameters of the object, respectively. Pose estimation is utilized to establish the relationship between the image space and physical space. Subsequently, several groups of comparative experiments are conducted with manual measurements as a reference. The experimental results indicate that, for log diameter measurement, approximately 97.33 % of the deviations between device measurement and conventional measurement can be expected to fall within a range of −0.54 cm and 0.61 cm, with mean absolute deviation (MAD) and mean relative deviation (MRD) of 0.25 cm and 0.78 %, respectively. In terms of volume estimations, MAD and MRD are found to be 0.23 and 0.56 %, respectively. The results imply that the device contributes to substantial improvements in accuracy, reliability, and traceability, and it shows promising potential to replace manual measurements.

Phenotyping methodologies of log end splitting in eucalyptus (Eucalyptus spp.)

AbstractThis study addresses the crucial consideration of log end splitting in breeding programmes for treated wood. There is a paucity of research focused on efficiently optimizing the phenotyping process for this particular trait. The study aimed to compare methodologies for log end splitting phenotyping and develop an image‐based crack evaluation approach. Initially, 32 eucalyptus clones underwent phenotyping using manual measurement, digital image analysis and visual evaluation. Results showed similar phenotypic values, but image analysis demonstrated better clone discrimination, reducing evaluation time to 78 h compared to manual measurement. The second part focused on testing convolutional neural network architectures (UNet, LinkNet and FPN) using real and synthetic images. U‐Net exhibited slight superiority based on higher Intersection over Union (IoU) values, exhibiting a high correlation (.89) with true values. This approach significantly reduced evaluation time to approximately 10.15 h, emphasizing its efficiency compared to traditional methods.

Log cross section quality metrics: Assessing the usability of roundwood image data for roundwood tracking

The proof of origin of roundwood is becoming increasingly important. As part of the digitization of the forest-based sector and to combat illegal logging, there is a growing interest in tracking wood logs individually. There are several previous publications on biometric wood log tracking based on log image data. A major problem for wood log tracking is the often poor visibility of discriminative features in log ends, such as the annual ring pattern, which makes it very difficult to track such logs. In this work we propose a method that quantifies the usability of image data for the purpose of roundwood tracking. There are no previous publications on this topic. More specifically, we propose a quality map that assesses the usability of different regions of the log cross section and a global quality metric that assesses the quality of entire images. We will show that the local and global quality metrics reliably assess the quality of log cross section images. Furthermore, we will show that the quality map can be used to improve the results of deep learning based log recognition systems. By combining the image data with the local quality map, the roundwood tracking system gets additional information about the usability of the different regions of the log cross section for roundwood tracking, which leads to improvements in most of the results.

Comparison of Modern Deep Neural Networks Architectures for Cross-section Segmentation in Images of Log Ends

Comparison of Modern Deep Neural Networks Architectures for Cross-section Segmentation in Images of Log Ends

Design and Implementation of an Intelligent Log Diameter Grading and Sorting Line Based on Machine Vision

In order to address the challenges posed by elevated manual labor costs and limited automation in traditional log diameter grading and sorting processes, this paper centers on the design and research of an intelligent log diameter grading and sorting line utilizing machine vision. The study focuses on logs with smaller diameters located in Fujian province, China. By analyzing production requirements, the study formulates the structure of the feeding, alignment, detection, and sorting zones to fulfill sorting functions. Using the YOLOv5 model, the system achieves accurate log end face positioning, and the diameter is computed through a designated algorithm. The operational process of the system is examined, and the control logic governing the production line is elucidated. Evaluating the practical performance of the production line, the study assesses the accuracy of diameter recognition, precision in grading, and operational efficiency. The results reveal that the absolute error in diameter detection for the sorting line averages 1.12 mm, with sorting accuracy exceeding 95%. The sorting line can automatically categorize logs with diameters ranging from 60 mm to 300 mm and lengths ranging from 2 m to 6 m, achieving an annual sorting capacity of 120,000 to 130,000 cubic meters. The research findings illustrate that the system fulfills the industry’s demands for log diameter grading and sorting, thereby enhancing economic efficiency for enterprises.

Using Improved YOLOv5 Model to Detect Volume for Logs in Log Farms

<p>In this paper, we propose a new computer vision model called SE-YOLOv5-SPD for counting the number of log ends in large wood piles in log farms. This task traditionally requires a lot of manpower and previous computer vision methods struggle to detect logs in low pixels and small objects in images. Our model is based on the YOLOv5 model and incorporates the Squeeze-and-Excitation Networks (SENet) attention module and SPD-Conv (Space-to-Depth Convolution) module to improve accuracy. We also compare the performance of the SE attention module and SPD-Conv module to the CBAM attention module and Focus module using the SE-YOLOv5-SPD model. Results show that the SE-YOLOv5-SPD model can achieve excellent results of 0.652 in mAP50:95, 0.912 in mAP50, 0.968 in Precision, and 0.864 in Recall in a low-resolution environment with interference, which is significantly better than other models. Our findings indicate that the SE-YOLOv5-SPD model is a promising solution for counting the number of log ends in wood piles.</p> <p> </p>

Characterization and Yield of Eucalyptus regnans F. Muell Logs for Lumber Production

The yield of Eucalyptus regnans logs for lumber production was evaluated. Crack width and length at each log end were measured. Two log-cutting plans were used to obtain sawn lumber. The first plan (PA) considered logs with diameters varying from 28 to 40 cm, and in the second plan (PB), the log diameters ranged from 42 to 56 cm (PB). Lumber yield was determined using two log volume methods: the Japanese Agricultural Standards (JAS) and Smalian’s equation. The deformations of E. regnans lumber were measured. The Australian and Chilean standards were used to classify sawn lumber. The results showed that logs had radial cracks at both log ends. Cracks were classified into two groups, considering the crack length. Regarding the lumber deformations, most boards exhibited level B bows and crooks in both cutting plans. Levels A and B twists were prevalent in PA, whereas in PB, level A significantly outnumbered level B. The lumber yield of E. regnans in PB was higher than in PA. The lumber yield determined by Smalian’s equation was higher than that determined by the JAS method. This research provides insight into the characterization of E. regnans for lumber production, highlighting its relevance in the forestry industry.

Can diameter and log end splittings be a good indicator of board end splitting in 21-year-old Eucalyptus grandis Hill ex-Maiden?

Eucalyptus spp. are currently cultivated in Brazil. In this article we highlight Eucalyptus grandis with its potential for mechanical processing. After felling trees, the splitting that occurs at the end of logs and boards obtained during mechanical processing represents an indirect measure of the level of growth stress present in the living trees. Even though such splitting can cause significant losses in the use of wood, knowledge of this pattern is still sparse. The most suitable eucalyptus trees for the production of sawn wood can be selected by evaluating the effect of the release of growth stresses in the form of cracks. To address this gap, we herein evaluate the effects of end splitting in logs and boards, resulting from the release of growth stresses during the felling of trees and mechanical processing, as a function of the diameter of 48 Eucalyptus grandis Hill ex-Maiden trees aged 21 years old. Splitting at the ends of boards in the green and dry states have a low relationship with the diameter of the logs, but it is not related to the end splitting of the tree logs. A high positive relationship was found between dry board splitting and green board splitting. Therefore, it is possible to estimate with accuracy the evolution of green board splitting according to the drying of the wood. Boards lose, on average, 44 cm (15%) from cracks, or an average of 32 cm (11%) from growth stresses and 12 cm (4%) from drying stresses. Green and dry board splitting decreased significantly from pith to bark. The splitting of dry and green boards had no significant variation between the analyzed positions of the log along the height of the tree.

A Fast Instance Segmentation Technique for Log End Faces Based on Metric Learning

The diameter of the logs on a vehicle is a critical part of the logistics and transportation of logs. However, the manual size-checking method is inefficient and affects the efficiency of log transportation. The example segmentation methods can generate masks for each log end face, which helps automate the check gauge of logs and improve efficiency. The example segmentation model uses rectangle detection to identify each end face and then traverses the rectangular boxes for mask extraction. The traversal of rectangular boxes increases the time consumption of the model and lacks separate handling of the overlapping areas between rectangular boxes, which causes a decline in mask extraction accuracy. To address the above problems, we propose a fast instance segmentation method for further improving the efficiency and accuracy of log-checking diameter. The method uses a convolutional neural network to extract the mask image, rectangular frame prediction image, and embed the vector image from the input image. The mask image is used to extract the log end face region, and the rectangular frame prediction image generates an enveloping rectangular frame for each log, which in turn divides the log end face region into instances. For the overlapping regions of rectangular boxes, a metric learning paradigm is used to increase the embedding vector distance between pixels located in different logs and decrease the embedding vector distance between pixels of the same log, and finally the mask pixels of the overlapping regions of rectangular boxes are instantiated according to the pixel embedding vectors. This method avoids repeated calls to the contour extraction algorithm for each rectangular box and enables fine delineation of pixels in the overlapping rectangular box region. To verify the efficiency of the proposed algorithm, the log working pile is photographed in different scenes using a smartphone to obtain the end face recognition database and divide the training set, validation set, and test set according to 8:1:1. Secondly, the proposed model is used to obtain log end face masks, and the log end face ruler diameter is determined by an edge-fitting algorithm combined with a ruler. Finally, the practicality of the algorithm is evaluated by calculating the check-rule diameter error, running speed, and the error of wood volume calculation under different national standards. The proposed method has 91.2% and 50.2 FPS of mask extraction accuracy and running speed, respectively, which are faster and more accurate than the mainstream instance segmentation model. The relative error of the proposed method is −4.62% for the check-rule diameter and −4.25%, −5.02%, −6.32%, and −5.73% for the wood volume measurement under the Chinese, Russian, American, and Japanese raw wood volume calculation standards, respectively. Among them, the error of the calculated timber volume according to our standard is the smallest, which indicates that the model in this paper is more suitable for application in domestic log production operations.

Tree log identification using convolutional neural networks

The identification of individual tree logs along the wood procurement chain is a coveted goal within the forest industry. The tracing of logs from the sawmill back to the forest would support the legal and sustainable sourcing of wood, as well as increase the resource efficiency and value of harvested timber. In this work, using a dataset of thousands of Scots pine (Pinus sylvestris L.) log end images displaying varying perspectives, lighting, and aging effects, we develop and assess log identification methods based on deep convolutional neural networks. The estimated rank-1 accuracy of our final model on an independent test set of 99 logs is 84 and 91% when allowing for random rotations of the log ends and when keeping each log at approximately fixed orientation, respectively. We estimate the scaling of these methods up to a template pool size of 493 logs, which reveals a weak dependence of accuracy on pool size for logs at fixed orientation. The deep learning approach gives superior results to a classical local binary pattern method, and appears feasible in practice, assuming that pre-filtering of the log database can be leveraged depending on the use case and properties of the queried log image. We make our dataset publicly available.

Traceability and quality assessment of Norway spruce (Picea abies (L.) H.Karst.) logs: the TreeTrace_spruce database

Key messageThe TreeTrace_spruce database contains images and measurements of 100 Norway spruce (Picea abies (L.) H.Karst.) logs from Northeastern France, each about 4.5 m long. The image database includes RGB images of large and small ends of the logs and hyperspectral and computed tomography (CT) images of wood discs sampled at both log ends. The 100 logs were also fully X-ray scanned with a CT device for roundwoods and their top surface was scanned with a terrestrial LiDAR device. The measurements performed on discs include wood local density, growth ring widths and pith location. This database is complementary to another one (TreeTrace_Douglas) resulting from the same ANR project TreeTrace, but if the objectives are similar, the protocols and conditions of acquisition are not the same for these two databases. TreeTrace_spruce dataset is available at https://doi.org/10.57745/WKLTJI and associated metadata are available at https://metadata-afs.nancy.inra.fr/geonetwork/srv/fre/catalog.search#/metadata/cffee2f1-18e1-4b53-9f5b-6cc4c66f1cb8.

A Novel Wood Log Measurement Combined Mask R-CNN and Stereo Vision Camera

Wood logs need to be measured for size when passing through customs to verify their quantity and volume. Due to the large number of wood logs needs through customs, a fast and accurate measurement method is required. The traditional log measurement methods are inefficient, have significant errors in determining the long and short diameters of the wood, and are difficult to achieve fast measurements in complex wood stacking environments. We use a Mask R-CNN instance segmentation model to detect the contour of the wood log and employ a binocular stereo camera to measure the log diameter. A rotation search algorithm centered on the wood contour is proposed to find long and short diameters and to optimal log size according to the Chinese standard. The experiments show that the Mask R-CNN we trained obtains 0.796 average precision and 0.943 IOUmask, and the recognition rate of wood log ends reaches 98.2%. The average error of the short diameter of the measurement results is 5.7 mm, the average error of the long diameter is 7.19 mm, and the average error of the diameter of the wood is 5.3 mm.

Ant Colony Optimization for Estimating Pith Position on Images of Tree Log Ends

The pith location is one of the most important features to detect in order to determine the quality of wood. Indeed, it allows to extract other important features. In this paper, we address the problem of pith detection on images of wood cross-sections. Taking such images can be done at little cost and with a high resolution. However, contrary to computed tomographic images, digital images exhibit disturbances like sawing marks, dirt or ambient light variations which make difficult the image analysis. Few studies have focused on such images. Furthermore these studies do some prior segmentation or cropping before the detection. We propose an approach for estimating the pith location without any requirements. Our method is based on an ant colony optimization algorithm. It is a probabilistic approach for solving this task. We validate our algorithm on images of Douglas fir captured after harvesting. The efficiency of this algorithm has been demonstrated by performance comparisons with other approaches. Experiments show an accurate and fast estimation and the algorithm could be used in real time, at sawmill environment or in forest, with a smartphone.

Traceability and quality assessment of Douglas fir (Pseudotsuga menziesii (Mirb.) Franco) logs: the TreeTrace_Douglas database

Key messageThe TreeTrace_Douglas database includes images and measurements at several stages of the processing of Douglas fir logs, from sawmill logyard to machine grading and destructive testing of boards, and is suitable for research on quality assessment and traceability. A total of 52 long logs, 156 short logs, 208 wood discs, and 346 boards were analyzed. The image data includes RGB images of log ends and board ends, RGB images and CT slices of strips, and a set of images of the boards (RGB, laser, and X-rays) obtained with an industrial board grading machine. The measurements include wood density, growth ring widths, pith and board location in the logs, heartwood and sapwood areas, mechanical properties of each board obtained by vibratory and static testing, and visual grading of the boards. Dataset is available at https://doi.org/10.15454/YUNEGL and associated metadata are available at https://metadata-afs.nancy.inra.fr/geonetwork/srv/fre/catalog.search#/metadata/d9eef6e4-f195-41f4-b6c2-2ab46adc637e.

A detection approach for bundled log ends using K-median clustering and improved YOLOv4-Tiny network

In log trade, precise measuring log volume is directly related to the economic benefits. Usually, the length of the same batch of wood is relatively fixed, but the shape and size of log end are different. Therefore, the variability of the size, shape and density of log end, and the sophisticated scenarios lead to great challenge for log detection from bundled log ends images. In addition, the large number of logs in the bundled log ends image results in less pixels occupied by a single log, and it is essential to detect the small log ends. In order to accurately detect each log from bundled log ends image, a new approach of log detection based on K-median clustering and improved You Only Look Once (YOLO)v4-Tiny network has been developed. The specific methods are as follows: (1) In consideration of the characteristics of log size, shape and its distribution, the K-median clustering method is used to select a more appropriate size of multi-scale anchor boxes to achieve a more consistent detection boxes for the log ends; (2) The detection scales of YOLOv4-Tiny network are increased to three and the spatial pyramid pooling (SPP) module is added to enhance the ability of feature extraction for small targets, such as small log end; (3) The self-attention mechanism based on squeeze and excitation (SE) is inserted into the deep structure of the network, which automatically determines the threshold without relevant professional knowledge to eliminate the high-dimensional noise that may result in error gathering of the centers of log ends and predicted boxes. The Precision, Recall and F1-score of YOLOv4-Tiny in the experimental test set are 91.87%, 94.91% and 0.93 respectively, while the three indicators are improved to 93.97%, 95.34% and 0.95 respectively by using the proposed model. Moreover, the complete intersection over union (CIoU) loss of our model is 2.46 and is reduced by 51.48% compared to the YOLOv4-Tiny with 5.07, which means the predicted boxes of our model are closer to the target bounding boxes. Consequently, the experiment results demonstrate that the performance of the proposed approach is better than that of YOLOv4-Tiny network.

Quality traits of plantation Eucalyptus nitens logs impacting volume and value recovery of structural sawn boards

Plantations of Eucalyptus species are planted and grown worldwide for short rotations and with limited silvicultural treatments mostly to produce pulplogs for the pulp and paper industry. These resources could be used as raw material for construction timber, to support the increasing need of renewable resources from the building sector. To use fast-grown Eucalyptus logs as a source of sawn timber log grading standards are needed, which can be developed accounting for log characteristics impacting sawn timber recovery. This study aims to examine the quality of fast-grown Eucalyptus logs and relate relevant log quality traits to sawn timber characteristics. Wood quality and log characteristics of forty-nine fast-grown Eucalyptus logs and the characteristics and structural properties of 268 sawn boards milled from those logs were investigated. Significant differences were found in wood quality characteristics from logs sourced from different positions in the stem. However, sawn boards did not differ in their wood quality traits according to log position, which influenced only the amount and type of knots on the board surface and some structural properties. Moreover, log characteristics including volume, taper, log end splits and stiffness significantly impacted important board recovery traits. The results of this study show that log characteristics such as volume, taper, log end splits and stiffness should be accounted for in log grading standards seeking to grade fast-grown Eucalyptus resources for different product classes.