Olika former av nyhetsbrev har funnits i århundraden. På Redpill Linpros kontor i Karlstadpubliceras veckoliga nyhetsbrev i form av videoinspelningar. För att de anställda påföretaget snabbt ska kunna ta del av innehållet i dessa videoinspelningar är detta projektssyfte och mål att skapa en automatiserad process som sammanfattar dessa veckoliganyhetsbrev. Sammanfattningen ska sedan kunna tas del av genom att spela upp dennapå en animatronisk fisk gjord av plast med inbyggda högtalare. Projektet implementerasmed begränsad hårdvara och ingen data delas med tredje part. För att uppnå detta målundersöks och implementeras olika verktyg som använder sig av artificiell intelligenssamt så modifieras den animatroniska fisken. Resultatet av arbetet är en modulär processsom upptäcker publicering, transkriberar, sammanfattar och läser upp sammanfattningenav veckoliga nyhetsbrev. Uppläsningen kan sedan spelas upp på den förutnämndaanimatroniska fisken med ett enkelt knapptryck. / Various predecessors of newsletters have been around for centuries. At Redpill Linpro’soffice in Karlstad, weekly newsletters are published in the form of video recordings. Inorder for the employees to be able to quickly take part of the content of these videorecordings, the purpose and goal of this project is to create an automated process thatsummarizes these weekly newsletters. The summary is then shared by the possibility ofplaying it on an animatronic fish made of plastic with built-in speakers. The project isimplemented with limited hardware and no data is shared with third parties. To achievethis goal, various tools that use artificial intelligence are researched and implemented,and the animatronic fish is modified. The result of the work is a modular process thatdetects publication, transcribes, summarizes and reads the summary of weekly newsletters.The reading can then be played back on the aforementioned animatronic fish witha simple push of a button.

The brown marmorated stink bug (BMSB), Halyomorpha halys (Stål) (Hemiptera: Pentatomidae), is an invasive insect from Asia that has recently become a major pest of agricultural crops and a nuisance to home and business owners in the Mid-Atlantic U.S. Since 2010, H. halys has been reported in many vineyards in Virginia, but the pest significance in this crop is unknown. Sampling was conducted in four commercial vineyards across Virginia in 2012 and 2013 to study the seasonal phenology and pest status of H. halys in vineyards. Adults moved into vineyards as early as May and laid eggs usually on the undersurface of leaves, but occasionally on the berry or the rachis. Grapevines were an early season reproductive host for H. halys. A vineyard adjacent to a sub-urban area with homes and buildings in proximity had an early season peak of H. halys as compared to vineyards adjacent to woods. However, populations declined sharply in late season due to the possible movement of bugs to more preferable host plants such as soybean and corn. In contrast, H. halys was recorded throughout the grape growing period in a vineyard that was surrounded by forests. Significantly more H. halys were recorded from border than interior section of vineyards. A degree-day model suggested that there were enough degree-days to complete a generation of H. halys in Virginia vineyards. H. halys caused direct injury to the grape berries at veraison and pre-harvest berries. Injury expression in the veraison berry can be described as an appearance of a small necrotic spot at the site of the stylets insertion. The spot gradually increased in size and the berries became deformed. H. halys is an economic pest of raspberry, causing direct injury to the berries. Sampling of stink bugs in raspberry plantings in southwestern Virginia showed that the Euschistus species were the most abundant stink bugs in 2008, 2009, 2011 and 2012. However, H. halys became the most abundant in 2013. / Master of Science in Life Sciences

Field experiments (2007-2009) and laboratory bioassays (2009) tested the efficacy of insecticides with short pre-harvest intervals, caneberry cultivar susceptibility, and geranium toxicity for reducing Japanese beetle (JB) activity on primocane-bearing caneberries. Deltamethrin, chlorantraniliprole, bifenthrin, lime-alum, and thyme oil reduced JB activity in the field. Deltamethrin, chlorantraniliprole, acetamiprid, an azadirachtin and pyrethrin mixture, an azadirachtin and neem oil extract mixture, and an extract of Chenopodium ambrosioides reduced JB activity during the bioassays. "Prelude" had significantly more JB than "Anne", "Caroline", "Heritage", "Dinkum", or "Himbo Top" and "Prime-Jan" had significantly more JB than "Prime-Jim". Compared to certain cultivars, "Heritage", "Caroline", "Himbo Top", and "Prime-Jan" had higher percentages of injured fruit and "Autumn Bliss", "Heritage", and "Caroline" produced greater marketable and overall yields. "Prime-Jan" produced more overall yield than "Prime-Jim"; marketable yields from both blackberry cultivars were similar. Defoliation was significantly less for "Dinkum", "Caroline", "Heritage", and "Anne" than for "Prelude" in 2008 and significantly less for "Caroline" and "Anne" than "Prelude" or "Fall Gold" in 2009. In field tests, previous consumption of geraniums lessened raspberry defoliation by JB. Bioassays indicated that JB activity was only reduced if JB were continually exposed to geranium. Therefore, the efficacy of geranium as a trap crop for JB may be limited. The stink bug species within the caneberries were identified (2008-2009) and Euschistus servus (Say) made up 48.1 % of the overall species composition. Stink bug injury to ripening raspberries was identified as small holes between drupelets; stink bug excretions also ruined fruit. / Master of Science in Life Sciences

Learning to play the piano usually involves learning to read sheet music and many hours of prac- tice, which can be seen as a tough task for beginners. This project presents an innovative ap- proach to piano learning by integrating a Raspberry Pi, Arduino, and LED lights with a Casio CTK-550 keyboard. This system interprets MusicXML files and visually guides users by lighting up the keys they need to press in the correct order, providing a more intuitive and engaging learn- ing experience. Additionally, it has a playback mode that makes the piano play the chosen song while illuminating the keys played. This allows the user to hear the song being played correctly before using the interactive mode to play the entire song themselves. These modifications to the piano aim to simplify the learning process and ease in new piano players by removing the big initial barrier of understanding sheet music. The project practices sustainability by reusing components from old projects and also follows several UN Sustainable Development Goals. After a few iter- ations, there was a product able to fulfill the goals set in advance. Future improvements could include improved lighting precision, additional learning modes, and more user-friendly file transfer possibilities.

Syftet med denna studie är att undersöka hur bearbetningstiden hos Snort intrångsskyddssystem varierar mellan två olika processorer; ARM Cortex-A53 och Cortex-A7. CPU-användningen undersöktes även för att kontrollera om bearbetningstid är beroende av hur mycket CPU Snort använder. Denna studie ska ge kunskap om hur viktig en processor är för att Snort ska kunna prestera bra när det gäller bearbetningstid och CPU användning samt visa det uppenbara valet mellan Cortex-A53 och Cortex-A7 när man ska implementera Snort IPS. Med hjälp av litteratursökning konstruerades en experimentmiljö för att kunna ge svar på studiens frågeställningar. Snort kan klassificeras som CPU-bunden vilket innebär att systemet är beroende av en snabb processor. I detta sammanhang innebär en snabb processor gör att Snort hinner bearbeta den mängd nätverkstrafik den får, annars kan trafiken passera utan att den inspekteras vilket kan skada enheten som är skyddat av Snort. Studiens resultat visar att bearbetningstiden i Snort på Cortex-A53 och Cortex-A7 skiljer sig åt och en tydlig skillnad i CPU-användning mellan processorerna observerades. Studien visar även kopplingen mellan bearbetningstiden och CPUanvändning hos Snort. Studiens slutsats är att ARM Cortex-A53 har bättre prestanda vid användning av Snort IPS avseende bearbetningstid och CPU-användning, där Cortex-A53 har 10 sekunder kortare bearbetningstid och använder 2,87 gånger mindre CPU. / The purpose of this study is to examine how the processing time of the Snort intrusion prevention system varies on two different processors; ARM Cortex-A53 and CortexA7. CPU usage was also examined to check if processing time depends on how much CPU Snort uses. This study will provide knowledge about how important a processor is for Snort to be able to perform well in terms of processing time and CPU usage. This knowledge will help choosing between Cortex-A53 and Cortex-A7 when implementing Snort IPS. To achieve the purpose of the study a literature search has been done to design an experimental environment. Snort can be classified as CPU-bound, which means that the system is dependent on a fast processor. In this context, a fast processor means that Snort is given enough time to process the amount of traffic it receives, otherwise the traffic can pass through without it being inspected, which can be harmful to the device that is protected by Snort. The results of the study show that the processing time in Snort on Cortex-A53 and Cortex-A7 differs and an obvious difference in CPU usage between the processors is shown. The study also presents the connection between processing time and CPU usage for Snort. In conclusion, ARM Cortex-A53 has better performance when using Snort IPS in terms of processing time and CPU usage, Cortex-A53 has 10 seconds less processing time and uses 2,87 times less CPU.

This thesis describes how a solution can be built to detect human flu-like symptoms. Flu-like symptoms are important to detect to prevent Covid-19 [6]. As people are returning to work there is a need for a simple way of detecting flu-like symptoms to prevent the spread of Covid-19. Other than a solution, this thesis concluded how human flu-like symptoms can be detected, with cameras specifically. This is to know what symptoms are most likely to work for a prototype. The technique of cameras and thermal cameras made this project possible as well as the technique of a single-board computer. The technique of cloud-based services is also an important part of this project. This project has resulted in a novel prototype using a single-board computer, cameras, and various cloud-based services to detect and inform a person if he or she has a human flu-like symptom.

This thesis deals with developing of image processing algorithm for robots controlled by informations taken from visual system. Core of the used system constitutes low-budget platform Raspberry Pi. Before the development of algorithm there is a series of test for image processing which discovers possibilities of used platform. Problem solution is divided to several parts. Limited performance frequently leads to individual problem solving. Afterall is shown sensitivity and performance analysis of developed solution.

Syftet med detta examensarbete var att skapa ett serverprogram i en enkortsdator som arbetar i realtid för att kunna hjälpa människor att köra mer ekonomiskt och miljövänligt i deras vardag. Detta var ett av målen ställda av produktbeställaren Ziggy Creative Colony. Ett mer långsiktigt mål från beställaren är att datorn skall installeras i en bil och kopplas till bilens on-board diagnostic-II (OBD-II)-uttag. Datorn ska sedan, via OBD-II, kunna samla information som till exempel hastighet, acceleration och bränsleflöde från bilens engine control unit (ECU). Serverprogrammet ska bearbeta denna information som sedan kommuniceras och visualiseras till bilföraren via en native mobilapplikation. Serverprogrammet byggdes i en linuxbaserad dator: Raspberry Pi och döptes av oss till Car-Pi. Car-Pi designades enligt arkitekturmönstret Model-View-Controller (MVC) som gör det lätt att underhålla, vidare-utveckla och implementera programmet av produktägaren, Ziggy Creative Colony, i framtiden. I och med denna rapport levererar vi programmet Car-Pi tillsammans med ett arkitektdokument och en prototyp till en Android mobilapplikation för att kunna testa Car-Pi och se hur det fungerar i verkligheten. / The purpose of this bachelor degree project was to create a server program in a single-board computer that will function in real time in order to help people drive more economically and eco-friendly. This was one of the goals set by the project owner, Ziggy Creative Colony. Another more long-term goal from Ziggy Creative Colony is that the computer should be installed in a car and connected to the car’s on-board diagnostics (OBD-II) connector. The computer should, via the OBD-II connection, be able to collect information such as speed, acceler-ation and fuel flow from the car’s engine control unit (ECU). The server program will then process this information that will later be communi-cated and visualized to the car driver via an Android mobile applica-tion. The server program was built in a linux-based computer: Raspberry Pi and named Car-Pi. Car-Pi was designed according to the well-known architectural pattern Model-View-Controller (MVC), which makes it easy to maintain, develop and implement the program by the project owner Ziggy Creative Colony in the future. With this report we are delivering the program Car-Pi together with an architectural document and an Android mobile application-prototype to show how Car-Pi functions in real life.

This master thesis project was carried in collaboration with Keolis AB. One of the company’s goals is to seek for a possible replacement to expensive custom hardware units by cheap single board computers. As a test case, a simple application is proposed, which implements driver identification by scanning the barcode of driving license (personnummer). The main objective of this project is to find a suitable single board computer, to implement the proposed driver identification application, to send the driving license number along with the timestamp in a web page and test the functionality of the single board computer according to procedures specified in ISO and IEC standards for road vehicles. A parser that analyzes the input string of a barcode reader was implemented in C programming language. The barcode reader scans a barcode or a QR code and the parser returns the content of the barcode symbol in ASCII character format. The driver license number as well as the timestamp should be published to a web page. A webpage was created using the Django Web Framework, which is a database-driven website. Each time a driving license barcode is scanned, a POST Http request method is performed and both the driving license and timestamp are stored in a SQLite database. Each time a GET request method is performed the data stored in the SQLite database is retrieved and presented in the website. The communication between the single board computer Raspberry Pi and the Django framework is achieved using cURL, which is an open source command line tool and library for transferring data with URL syntax. The data of the website will be manipulated in the backend. Moreover, heat and humidity environmental testing were performed as described in ISO and IEC standards for road vehicles, to evaluate the functionality of the system under certain environmental conditions. These tests showed the working temperature range and the humidity range that the Raspberry Pi can tolerate. As a conclusion, it can be stated that Raspberry Pi can be used in the passenger compartment with expected temperatures to be below 100 Celsius, but not in the engine compartment where temperatures more than 100 Celsius can occur. In addition, Raspberry Pi can perform in all the levels of humidity that has been tested. If it is necessary to be employed in other bus compartment with increased temperature, a more expensive robust embedded single board Linux computer should be chosen. Future work should include vibrations and immunity testing, in order to fully qualify with the ISO and IEC standards. These types of tests are costly and should therefore be performed by automotive manufacturers or other parties who are expected to bear such a cost. / Detta examensarbete genomfördes i samarbete med Keolis Sverige AB. Ett av företagets mål är att söka efter möjliga system som ersättning till dyra, specialanpassade hårdvaruenheter och istället övergå till billiga enkortsdatorer. Som ett testfall föreslås ett enkelt program, som genomför identifiering av föraren genom att skanna streckkoden på körkortet (personnummer). Huvudsyftet med projektet är att hitta en lämplig enkortsdator, att implementera den föreslagna föraridentifieringsapplikationen, skicka körkortsnumret/personnumret med tidsstämpel till en webbsida och testa funktionaliteten hos enkortsdator enligt testrutiner som beskrivs av ISO- och IEC-standarder för vägfordon. En parser som analyserar indatasträngen av en streckkodsläsare implementerades i programmeringsspråket C. Streckkodsläsaren skannar en streckkod eller en QR-kod och parsern returnerar innehållet i streckkoden på ASCIIteckenformat. Körkortsnumret samt tidsstämpel publiceras på en webbsida. En webbsida har skapats med Django Web Framework, som är en databasdriven webbplats. Varje gång körkortets streckkod skannas, skickas en POST http-begäransmetod som utförs varvid både körkort och tidsstämpel lagras i en SQLite databas. Varje gång en GET-begäran skickas, lagras data i SQLite databasen och presenteras på webbplatsen. Kommunikationen mellan enkortsdatorn Raspberry Pi och ett Django-ramverk uppnås med hjälp av cURL, som är ett kommandoradsverktyg med öppen källkod, och ett bibliotek för att överföra data med URL-syntax. Uppgifterna på webbplatsen manipuleras i backend. Miljötålighetsprovning med avseende på värme- och fuktighet har utförts för att utvärdera systemets funktionalitet under specifika miljöförhållanden. Testerna specificeras i ISO- och IEC-standarder för vägfordon. Dessa tester visade vilka arbetstemperaturer och vilken luftfuktighet som Raspberry Pi klarar. Det kan konstateras att Raspberry Pi kan användas i passagerarutrymmet, där temperaturen förväntas ligga under 100 Celsius, men inte i motorrummet där temperaturer högre än 100 Celsius kan förekomma. Vidare har Raspberry Pi visat sig fungera vid de nivåer av luftfuktighet som har förkommit i testerna. I de fall där systemet skall användas i miljöer med högre temperaturer bör en dyrare och mer robust inbyggd (embedded) Linux-enkortsdator väljas. Det framtida arbetet bör omfatta vibrations- och elstörningstester för att fullt ut säkerställa att systemet klarar gällande ISO- och IEC-standarder. Dessa typer av test är kostsamma och bör därför genomföras av fordonstillverkare eller andra aktörer som förväntas kunna bära en sådan kostnad.

In recent years, AI has developed significantly and become more popular than ever before. The applications of AI are expanding, making knowledge about its application and the systems it can be applied to more important. This project compares and evaluates deep learning models for object detection of traffic signs on the embedded systems Nvidia Jetson Nano and Raspberry Pi 3 Model B. The project compares and evaluates the models YOLOv5, SSD Mobilenet V1, FOMO, and Efficientdet-lite0. The project evaluates the performance of these models on the aforementioned embedded systems, measuring metrics such as CPU usage, FPS and RAM. Deep learning models are resource-intensive, and embedded systems have limited resources. Embedded systems often have different types of processor architectures than regular computers, which means that some frameworks and libraries may not be compatible. The results show that the tested systems are capable of object detection but with varying performance. Jetson Nano performs at a level we consider sufficiently high for use in production depending on the specific requirements. Raspberry Pi 3 performs at a level that may not be acceptable for real-time recognition of traffic signs. We see the greatest potential for Efficientdet-lite0 and YOLOv5 in recognizing traffic signs. The distance at which the models detect signs seems to be important for how many signs they find. For this reason, SSD MobileNet V1 is not recommended without further trai-ning despite its superior speed. YOLOv5 stood out as the model that detected signs at the longest distance and made the most detections overall. When considering all the results, we believe that Efficientdet-lite0 is the model that performs the best. / Under de senaste åren har AI utvecklats mycket och blivit mer populärt än någonsin. Tillämpningsområdena för AI ökar och därmed blir kunskap om hur det kan tillämpas och på vilka system viktigare. I det här projektet jämförs och utvärderas djupinlärningsmodeller för objektdetektering av trafikskyltar på de inbyggda systemen Nvidia Jetson Nano och Raspberry Pi 3 Model B. Modellerna som jämförs och utvärderas är YOLOv5, SSD Mobilenet V1, FOMO och Efficientdet-lite0. För varje modell mäts blandannat CPU-användning, FPS och RAM. Modeller för djupinlärning är resurskrävande och inbyggda system har begränsat med resurser. Inbyggda system har ofta andra typer av processorarkitekturer än en vanlig dator vilket gör att olika ramverk och andra bibliotek inte är kompatibla. Resultaten visar att de testade systemen klarar av objektdetektering med varierande prestation. Jetson Nano presterar på en nivå vi anser vara tillräckligt hög för användning i produktion beroende på hur hårda krav som ställs. Raspberry Pi 3 presterar på en nivå som möjligtvis inte är acceptabel för igenkänning av trafikskyltar i realtid. Vi ser störst potential för Efficientdet-lite0 och YOLOv5 för igenkänning av trafikskyltar. Hur långt avstånd modellerna upptäcker skyltar på verkar vara viktigt för hur många skyltar de hittar. Av den anledningen är SSD MobileNet V1 inte att rekommendera utan vidare träning trots sin överlägsna hastighet. YOLOv5 utmärkte sig som den som upptäckte skyltar på längst avstånd och som gjorde flest upptäckter totalt. När alla resultat vägs in anser vi dock att Efficientdet-lite0 är den modell som presterar bäst.