Java程序辅导

LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 1 
 
 
 
 
LAB 1 – FORESTER-IN-A-BOX PRODUCT DESCRIPTION 
 
Aaron Reynolds 
Old Dominion University 
CS411W 
Professor J. Brunelle 
February 14, 2022 
Final Version 
 
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 2 
 
Contents 
 
1. Introduction ........................................................................................................................... 4 
2. Forester-in-a-Box Product Description ............................................................................... 5 
2.1 Key Product Feature and Capabilities ......................................................................... 5 
2.2 Major Components ........................................................................................................ 7 
3. Identification of a Case Study............................................................................................. 10 
4. Forester-in-a-Box Product Prototype Description ........................................................... 11 
4.1 Prototype Architecture ................................................................................................ 11 
4.2 Prototype Features and Capabilities .......................................................................... 12 
4.3 Prototype Development Challenges ................................................................................. 14 
5. Glossary ................................................................................................................................ 16 
6. References ............................................................................................................................. 18 
 
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 3 
 
List of Figures 
Figure 1 – Forest Ownership in Virginia …………………………………………………….. 4 
Figure 2 – Key Features and Capabilites of Forester-in-a-Box …………………………….. 6 
Figure 3 – Forester-in-a-Box Major Functional Component Diagram ...………………….. 7 
Figure 4 – Sensor Node Diagram ……………………………………………………………... 8 
Figure 5 – Prototype Major Functional Component Diagram ……………………………. 12 
List of Tables 
Table 1 – Development Tools …………………………………………………………………. 9 
Table 2 – Real-World Product vs. Prototype ………………………………………………. 13 
 
 
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 4 
 
1. Introduction 
Forestry, the third leading industry in Virginia, generates many important products such as 
paper, lumber, and energy (Mayes, 2016). According to Figure 1, provided by the Virginia 
Department of Forestry, of all forest land in Virginia, families and individuals own 60% 
(Virginia Department of Forestry, 2021). Typically, private woodlot owners implement a set it 
and forget it management plan resulting in an unhealthy forest with diminishing returns to the 
owner.  
Figure 1 
Forest Ownership in Virginia 
 
Private woodlot owners have a limited understanding of forestry and their woodlot’s 
important characteristics. Combine this with an inability to consistently monitor health and 
growth - because of the difficulty of appropriately timing forest management operations, or an 
unwillingness to pay for the expenses of professional forester help - and private woodlot owners 
end up with a minimally managed forest that produces a minimal monetary yield. Private 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 5 
 
woodlot owners need an inexpensive, simple, and comprehensive solution to monitor and 
evaluate their woodlots to maximize value, improve their management decisions, and prevent 
losses. 
Forester-in-a-Box is a hardware/software solution aimed at providing the private woodlot 
owner with timely forest management recommendations based on sensor data gathered in their 
woodlot. With the use of Forester-in-a-Box, the private woodlot owner would obtain a better 
understanding and evaluation of their woodlot, constant health and growth monitoring, and 
professional-level recommendations on woodlot management. Forester-in-a-Box allows the 
woodlot owner to manage their forest without paying a forester for periodic visits and 
recommendations which result in requiring the private woodlot owner to share their profit. 
2. Forester-in-a-Box Product Description 
Forester-in-a-Box is a full suite of hardware and software tools for monitoring pine woodlots 
to provide timely forest management recommendations, health alerts, and volume and value 
estimates to maximize the health and value of the private woodlot owner’s woodlot. Forester-in-
a-Box utilizes a sensor network throughout the woodlot to monitor growth and health. 
2.1 Key Product Features and Capabilities 
Forester-in-a-Box provides a comprehensive solution based on the specific characteristics of 
the private woodlot owner’s woodlot - starting with sensor nodes that are to be placed in the 
forest for data collection, and a planning tool for hardware and sensor setup recommendations. 
After the initial setup and data collection begins, Forester-in-a-Box’s sensor nodes can transfer 
data to a backend server via a cellular connection or a mobile application which can be accessed 
from a user’s mobile device. 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 6 
 
An important key feature of Forester-in-a-Box is the easy-to-use web application. The web 
application provides the user with access to information such as Growth Forecasting, Value 
Estimations, Health Alerts, Management Recommendations, and an accessible database with 
previous images of the woodlot owner’s woodlot. The Growth Forecasting feature allows access 
to the projected growth and health of a woodlot owner’s woodlot in the form of a chart. Value 
Estimations provides the woodlot owner with lumber prices and a value estimate for their 
woodlot. Health Alerts assesses the health of the woodlot and give the owner alerts when 
diseases and pests are noticed, or overall health is low. Management Recommendations informs 
the woodlot owner on how to treat diseases, kill pests, and when to thin or harvest their woodlot. 
Lastly, the gallery provides a timeline of photos of an owner’s woodlot. The key features and 
capabilities provided by Forester-in-a-Box are shown in Figure 2. 
Figure 2  
Key Features and Capabilities of Forester-in-a-Box 
 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 7 
 
2.2 Major Components (Hardware/Software) 
The Major Functional Component Diagram, shown in Figure 3 identifies the major hardware 
and software components of Forester-in-a-Box. A top-level view of the information flow is also 
illustrated. The diagram begins with a web of sensors gathering images in three directions 
horizontally and one vertically. These sensors then share data collected across the network to 
promote redundancy and for easy upload from any sensor node. Data upload is handled either 
manually through a mobile application - accessed on a user’s mobile device - or by cellular 
connection.  
Figure 3 
Major Functional Components Diagram 
 
When the information reaches the backend server, it is processed to produce health alerts, 
growth forecasting, value estimations, and management recommendations that the user can view 
through a web application which can be accessed by web browser on a personal computer or 
mobile device. 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 8 
 
Forester-in-a-Box’s hardware package, shown in Figure 4, consists of a Raspberry Pi with 
four IR-Cut toggle camera modules, a long-range wireless module, a Sleepy Pi module, and - 
upon user request - a cellular modem. Each sensor package is configured with these components, 
which can be paired with several other Forester-in-a-Box’s sensors to create a sensor network.  
Figure 4 
Sensor Node Diagram 
 
The camera modules included are for taking still images of the trunks and canopy to 
evaluate health and growth. The long-range wireless module allows the devices to create an ad-
hoc network for coordination and redundancy. The Sleepy Pi module will allow the device to go 
into a sleep mode for efficient power usage over time. Finally, the optional cellular modem that 
can be included will eliminate the need for the mobile application by providing a continuous 
real-time upload of data to the Forester-in-a-Box servers for calculations; however, this option is 
only plausible for areas with reliable cellular service. Other major components include a mobile 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 9 
 
device and personal computer with internet access where the user can access either the mobile 
application or web application. 
Forester-in-a-Box’s software components, shown in Table 1, consists of ForestCrowns, 
OpenCV, TensorFlow, and ArcGIS. Forester-in-a-Box will use the ForestCrowns software tool 
to aid in determining canopy cover and canopy health. OpenCV is an open-source computer 
library used for image processing and performing computer vision tasks such as detecting tree 
trunks. TensorFlow is an end-to-end machine learning platform that will aid in training the 
neural network used to determine whether diseases and pests are present. Lastly, ArcGIS is 
mapping and analytics software that Forester-in-a-Box will use to help find the optimal locations 
for sensors and create heatmaps to aid the user in understanding where problems are present in 
their woodlot.  
Table 1 
Development Tools 
 
 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 10 
 
Other software development components include GitLab, Amazon RDS, and MySQL. Gitlab 
is used for version control and issue tracking. Forester-in-a-Box stores the user’s woodlot data on 
the backend server using Amazon RDS and MySQL. The software configuration is developed 
using Java, Python, C++, HTML, CSS, and JavaScript to provide efficient computation and an 
easy-to-use user interface. 
3. Identification of a Case Study 
The primary intended users of Forester-in-a-Box are private woodlot owners who own a ten 
or more acre woodlot. Private woodlot owners end up with a minimally managed forest due to 
the difficulty of appropriately timing forest management operations; or an unwillingness to pay 
for the expenses of professional forester help. After the initial setup, Forester-in-a-Box will help 
private woodlot owners monitor and evaluate their woodlots to maximize value, improve their 
management decisions, and prevent losses.  
 Public foresters and consulting foresters are secondary users of Forester-in-a-Box. With 
Forester-in-a-Box, foresters can maintain multiple woodlots less demanding than before by 
eliminating manual data gathering thus, reducing the time taken to monitor a woodlot's health.  
 A case study group consists of private woodlot owners and foresters in Virginia who own 
and maintain pine woodlots. The private woodlot owners would provide feedback on how 
Forester-in-a-Box is impacting the health and maintenance of their woodlot. The foresters of this 
group will report on how Forester-in-a-Box is influencing their management of multiple 
woodlots. 
 As Forester-in-a-Box continues to grow, the user base would expand from private 
woodlot owners, public foresters, and consulting foresters to include lumber companies, non-
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 11 
 
profit environmental organizations, and state and federal environmental organizations. Lumber 
companies will better understand the health and population of a woodlot without a manual 
assessment of the land plot, reducing the difficulty of thinning operations and tree selection. 
Environmental organizations would use the Forester-in-a-Box database of forest health and 
growth forecasting from woodlots for research and analysis. 
4. Forester-in-a-Box Product Prototype Description 
The Forester-in-a-Box Prototype focuses on the software implementation of the web 
application where the user can create and manage an account related to the private woodlot. The 
features of the web application - growth profile, value estimation, health alerts, and management 
recommendations will remain largely intact. To achieve the key prototype objective of 
demonstrating the use of data from monitoring the growth timeline of a woodlot to help the 
owner manage and improve their property, the data that the sensor network would collect over 
time is a fabrication of actual data for the prototype implementation. 
4.1 Prototype Architecture (Hardware/Software) 
The Forester-in-a-Box prototype's architecture, shown in Figure 5, is mostly the same as the 
real-world product. The primary difference is the elimination of the sensor network hardware and 
the use of simulated data for the sensor network output. Forester-in-a-Box runs on Amazon Web 
Services Elastic Beanstalk and uses Amazon Relational Databases with MySQL. The web 
application runs on personal computers and mobile devices with web capabilities and access to a 
web browser. The web browser is where users will be able to access the main functionality of the 
Forester-in-a-Box software. The prototype differs by removing sensor nodes and the mobile 
application.  
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 12 
 
Figure 5 
Prototype Major Functional Component Diagram 
 
4.2 Prototype Features and Capabilities 
The features that the Forester-in-a-Box prototype implements are in Table 2. The sensor 
network will not be a part of the prototype, and Forester-in-a-Box will simulate this data. The 
prototype will implement most features completely; however, there will be partial 
implementations due to various reasons. The key features include the planning tool, forest health 
and growth tracking, management recommendations, growth projections, and value estimations. 
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 13 
 
Table 2 
Real-World Product vs. Prototype 
 
 The prototype will partially implement the planning tool. The account creation portion 
includes obtaining coordinates for the woodlot location which we will use to display a map to the 
user with ArcGIS. The prototype will not recommend the optimal placement for a cellular hub. 
However, the prototype will still give a sensor node count recommendation based on area and 
optimal manual hub location based on the owner's home location. The owner will provide the 
boundaries of their woodlot, and the prototype will recommend sensor node locations based on 
those boundaries. The sensor node location recommendations will be based on ArcGIS 
topographic map technology. 
 Forest health and growth tracking include tracking canopy closure, canopy color, stem 
growth, and stem quality. The Forester-in-a-Box prototype will track canopy closure, color, and 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 14 
 
stem growth utilizing OpenCV and TensorFlow. Forester-in-a-Box's prototype will recognize 
when the canopy is overgrown or there are discolorations in the canopy. This will be done using 
photos found on the internet or taken by the development team. Tracking stem growth will be 
accomplished by pixel calculations on photographs of the tree's trunk. The prototype will 
partially implement the stem quality feature by only identifying obvious issues such as changes 
in shape and color rather than pinpointing pests and diseases. 
 Forester-in-a-Box's prototype will fully implement the management recommendations 
features. The recommendations will include thinning, weed removal, harvesting, and health 
inspection based on growth and health data. Growth projections will be fully implemented and 
use a set of false data points run through an algorithm to project future growth. These points and 
data will be shown in a web-based excel chart and graph. Value estimations will also be fully 
implemented and rely on a web scraper to gather timber values from TimberUpdate to calculate 
the value of an owner's woodlot. 
4.3 Prototype Development Challenges 
Forester-in-a-Box is an intricate application with many interdependent parts. There are 
many challenges the development team faces and that will arise as the prototype gets further into 
development. One of the largest challenges will be gathering forestry images for testing machine 
learning and algorithms. While this problem is partially mitigated through the decision of using 
false data for growth projection and value estimation charts, the development team still needs to 
gather enough photos of trees of different ages to showcase the computer vision side of Forester-
in-a-Box.  
           Another challenge is using technology that is foreign to the development team - AWS, 
ArcGIS, TensorFlow, OpenCV, and possibly Docker. Each part plays a significant role in the 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 15 
 
main functionality of Forester-in-a-Box. With the time constraints of one semester, learning 
everything and implementing it all will be a challenge. Further still is developing the algorithms 
for each specific working part and getting everything to work along with each other seamlessly.  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 16 
 
5. Glossary 
 
• Basal Area:  The cross-sectional area of a tree at breast-height. A common way to 
describe stand density.  
• Canopy: The upper level of a forest dominated by the foliage of the forest’s trees. 
• Canopy cover: Measure of land area covered by forest canopy. 
• Crown/Canopy closure: Crown closure, also known as canopy closure, is an integrated 
measure of the canopy over a segment of the sky hemisphere above one point on the 
ground.  
• Cruise:  A sample measurement of a stand used to estimate the amount of standing 
timber that the forest contains. 
• Dendrometer: Device for measuring trees.  
• Diameter at Breast-Height (DBH): A standard measurement of a tree’s diameter taken 
at 4.5 feet from its base. 
• Forest Management: Catchall term for various tasks that could be performed to promote 
forest health and productively broadly. 
o Thinning: Reduces canopy crowding and promotes faster growth of the remaining 
trees. This can be practiced in several ways, such as systematically removing 
every fourth tree. Depending on the age, condition, and composition of the forest, 
commercial products may be harvested during a thinning operation. 
o Harvesting: Partial or complete removal of trees in a stand. 
o Other management practices may include pruning, pesticide application, and 
various timber stand improvements. 
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 17 
 
• Growth region: A category of a woodlot which has similar growing conditions due to 
slope aspect and slope percentage. These can be treated as a subdivision of the woodlot 
for sampling purposes. 
• Hectare:  Metric unit of area equal to a square with 100m sides. One hectare contains 
about 2.47 acres.  
• Softwood Trees: Needle-bearing trees, gymnosperms; includes pines, spruce, fir, etc. 
• Stand: A contiguous community of trees sufficiently uniform in composition, structure, 
age, size, class, distribution, spatial arrangement, site quality, condition, or location to 
distinguish it from adjacent communities. 
• Timber stand improvements: Forest management practices that improve the vigor, 
stocking, composition, productivity, and quality of forest stands. Improvement results 
from removing lower quality trees and allowing crop trees to fully use the growing space. 
The chief aim of Timber stand improvements is continued production of more and better 
timber products. These practices can be used to convert lower quality stands into faster 
growing and more productive forests of desirable species. 
 
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 18 
 
6. References 
 
Rochester, M. (n.d.). Forest management. Maine Forestry. Retrieved December 5, 2021,  
from https://www.maineforestry.net/forest-management.  
Basic inventory calculations. (n.d.). Rural Tech. Retrieved December 5, 2021,  
from http://www.ruraltech.org/virtual_cruiser/lessons/lesson_10/Lesson_10_PDF.pdf. 
Brandeis, T. J., Hartsell, A. J., Randolph, K. C., & Oswalt, C. M. (2018, September).  
Virginia’s forests, 2016 (SRS-223). United States Department of Agriculture. 
https://dof.virginia.gov/wp-content/uploads/USFS-SRS-223-VAs-
Forests2016_outpub.pdf.  
Dale, V. & D., T.W. & Shugart, Herman. (1985). A Comparison of Tree Growth  
Models. Ecological Modelling. 29. 145-169. 10.1016/0304-3800(85)90051-1  
Faulkena, J. (2021, October 15). Interview with a State Forester (D. Mays, Interviewer) 
 [Review of Interview with a state forester].  
Fiala, A. C., Garman, S. L., & Gray, A. N. (2006). Comparison of five canopy cover estimation 
techniques in the western Oregon Cascades. Forest Ecology and Management, 
 232(1–3), 188–197. https://doi.org/10.1016/j.foreco.2006.05.069  
Forest products and markets. (n.d.). Virginia Department of Forestry. Retrieved December 5,  
2021, from https://dof.virginia.gov/forest-markets-sustainability/forest-products-markets/. 
Forest resource information. (n.d.). Virginia Department of Forestry. Retrieved December 5, 
2021, from https://dof.virginia.gov/forest-markets-sustainability/forest- 
inventory/forestresource-information/. 
  
LAB 1 FORESTER-IN-A-BOX PRODUCT DESCRIPTION 19 
 
Growing loblolly pine for profit (No. 33). (2018, March). Virginia Department of Forestry. 
 https://dof.virginia.gov/wp-content/uploads/FT0033-Growing-Pine-for-Profit_pub.pdf. 
Jennings, S. (1999). Assessing forest canopies and understorey illumination: canopy closure, 
canopy cover and other measures. Forestry, 72(1), 59–74. 
https://doi.org/10.1093/forestry/72.1.59.  
Mayes, F. (2016, May 25). Southern states lead growth in biomass electricity generation. US 
Energy Information Administration. Retrieved December 5, 2021, from 
https://www.eia.gov/todayinenergy/detail.php?id=26392.  
Self, B. (2019, October). Timber stand improvement (No. 1281). Mississippi State University 
Extension. From http://extension.msstate.edu/publications/timber-stand-improvement. 
Southern pine beetle: When beetles attack. (n.d.). American Tree Farm System. Retrieved 
December 5, 2021, from https://www.treefarmsystem.org/SPBwhenbeetlesattack.  
Stelzer, H. E. (2015, March). Selling timber: What the landowner needs to know (No. G5051). 
University of Missouri Extension. 
https://www.ncforestservice.gov/Urban/pdf/UMO_LO_selling_timber.pdf. 
The value of pine production. (2014, June). Virginia Department of Forestry. From  
https://dof.virginia.gov/wp-content/uploads/FT0021-The-Value-of-
PineProduction_pub.pdf. 
Why protect your pine forest? (2007, January). Virginia Department of Forestry. From 
https://dof.virginia.gov/wp-content/uploads/Protect-Your-Pine-Forest_pub.pdf.  
Winn, M. F., Palmer, A. J., L, S. M., & Araman, P. A. (2016, October). ForestCrowns: A 
transparency estimation tool for digital photographs of forest canopy’s (SRS-215). United 
States Department of Agriculture. https://www.srs.fs.usda.gov/pubs/gtr/gtr_srs215.pd