Additionally, the 36-month beta value for UGI is 1.12. There are mixed opinions on the stock, with 2 analysts rating it as a “buy,” 0 rating it as “overweight,” 1 rating it as “hold,” and 0 rating it as “sell.” The average price predicted ...

UGI Corp. (UGI) vs. Its Peers: A Comparison

Stocks: Real-time U.S. stock quotes reflect trades reported through Nasdaq only; comprehensive quotes and volume reflect trading in all markets and are delayed at least 15 minutes. International stock quotes are delayed as per exchange requirements ...

UGI Corp.

U.S. Reps. Chrissy Houlahan and Dan Meuser have introduced legislation to eliminate the problematic piping that caused the blast.

West Reading chocolate factory explosion: Lawmakers propose eliminating problematic pipe linked to blast

VALLEY FORGE, Pa., October 12, 2023--(BUSINESS WIRE)--UGI Corporation (NYSE: UGI) announced today that Jason Rich has been appointed Vice President and Treasurer, effective immediately.

UGI Corporation Appoints Jason Rich as Vice President and Treasurer

Investors might want to bet on UGI (UGI), as it has been recently upgraded to a Zacks Rank #2 (Buy). This rating change essentially reflects an upward trend in earnings estimates -- one of the ...

What Makes UGI (UGI) a New Buy Stock

The Pennsylvania Public Utilities Commission on Wednesday provided new information about an explosion last year that destroyed a Dauphin County home and injured two people. The PUC's Safety Division detailed the events it says led up to the blast in the 3200 block of Crest Road in Susquehanna Township:

PUC releases new information about explosion that destroyed Dauphin County home

On Dec. 12, 2022, Pronto Plumbing placed an emergency ticket with the Pennsylvania One Call system to alert utilities to locate and mark underground lines around the home. UGI Utilities did not locate and mark their line on Dec. 12.

PUC releases new information about 2022 house explosion

I nvestigators with Pennsylvania's Public Utility Commission (PUC) released a statement Wednesday regarding their investigation into a 2022 house explosion in suburban Harrisburg.

Officials release investigation update on home explosion in Susquehanna Twp

DAUPHIN COUNTY, Pa. (WHP) — Investigators with Pennsylvania's Public Utility Commission (PUC) released a statement Wednesday regarding their investigation into a 2022 house explosion in suburban Harrisburg.

Officials release update on investigation into Susquehanna Twp home explosion

We are at the cusp of a Cambrian explosion in user-generated content. It’s unlike anything we’ve seen before. Chances are, you’re already starting to see it all over social media; the ultra-realistic art, the deep fakes, the videos that look like moving paintings. Generative image content has steadily evolved from simple static images to more complex interactive visuals. New applications powered by AI are the force behind this trend, allowing users to create unique and dynamic visual experiences that can be used in a variety of contexts, such as advertising, entertainment, education, and more. AI-powered user generative image content (UGIC) has a wide array of applications. From more effective data visualization to creating virtual worlds, the exponential adoption of AI has removed all the technical barriers to entry for everyone. What was once reserved for technical specialists is now wholly accessible to anyone with a smartphone. How did all of this start? Where is it going? And what kind of role will artificial intelligence play in the evolution of generative image content? Introduction to Generative Image Content Generative image content is a subset of the broader machine learning field. The goal has been to program computers to emulate human thought processes and problem-solving. At the end of the day, this means computers can . learn without being programmed In 2014, image generation took a giant leap when were developed. This led to much more realistic images generated by AI. Now, generative image content can be used to create highly personalized and interactive visual experiences. generative adversarial networks (GANs) Generative images are created using algorithms, which are sets of instructions that allow computers to generate visuals from a given set of parameters. This is known as machine learning. The idea behind the current system of generative image AI has its roots in two different technological advances, one that came about in the 1950 and the other in the 1960s. Joseph Weisenbaum was the brilliant mind behind the , which was, in essence, the first chatbot. ELIZA program It was a simulation where the program learned from the response of the patient, with the program pretending to be a legitimate human psychologist. It was so good that many patients had no idea ELIZA wasn’t a real person. About a decade later, a process was being studied. It’s a complex system, even today, but it is the foundation for what is a much more common term today: Machine learning. Here’s the : known as deep learning official definition “Deep learning is a subset of AI and machine learning that uses multi-layered artificial neural networks to deliver state-of-the-art accuracy in tasks such as object detection, speech recognition, language translation, and others.” Without the above tech from decades ago, modern AI would not exist as we know it. Adding Artificial Intelligence Into the Mix The rise of artificial intelligence (AI) has had a major impact on the development of generative image content. AI algorithms can build more complex visuals and interact with user input in real time. This has allowed for some impressive applications, such as interactive virtual worlds enabling users to control their environment. The tech can also be used to create personalized visuals based on a user’s preferences or behavior. This allows for more immersive experiences in gaming, advertising, and even education. Now, platforms are leveraging this ability to offer users exciting new capabilities, and both users and investors are all for it. Lensa was one of the first to hit the mass market, and it was well rewarded. The app hit several astounding milestones almost immediately, including making , just after its initial launch. several million dollars within a single month Lensa: Starting It All While Lensa wasn’t the first of its kind, it demonstrated the demand for these types of AI tools. Lensa renders photos into artwork based on text prompts. Leveraging AI models, the app allows users to create virtually any kind of artwork imaginable, from cosmic fantasy to 80s-themed pop. And if that wasn’t impressive enough, the app also allows users to transform their selfies into highly customizable portraits. To achieve this, Lensa uses a copy of open-source code called Stable Diffusion to run its program. This provides access to billions of images from around the web and then compiles them into a dataset. Then, the AI uses these images to “learn” techniques that it applies to generate new works. So far, it’s been on social media, with tens of millions of downloads since it went live. But the app has drawn criticism from artists, claiming the art it generates is based on stolen work. a big hit So while Lensa’s future is uncertain, it has demonstrated a sliver of what is possible when combining images and AI. And with consumers demanding more, the race is on to see what all can be done with generative imagery.. The Future of AI Art With the help of algorithms and deep learning techniques, computers can now create original works of art that rival those of human artists. This disruption has the potential to shake up various industries that incorporate art, such as advertising, fashion, and entertainment. Interior decoration Some have a natural aptitude for interior design. They can visualize how a room can look with different colors, furniture, and art. For many, this can be quite challenging, but not for long. Generative AI has given way to platforms such as , which can act as your own virtual interior designer. Interior AI Create an image from (almost) nothing. While Lensa was able to edit and change existing photos, apps such as allows users to create ultra-realistic selfies using reference images or even just text. Users are able to create a completely new and unique photo from something as simple of a prompt as “a woman petting a dragon.” This begs the question, will photographers soon be out of a job? Artisse AI Virtual photoshoots Platforms such as could potentially make photoshoots obsolete as well. This website allows users to upload an image of themselves and then generate images in a variety of settings, poses, and more. What’s more, the images come out incredibly realistic, so convincing your friends you went to Paris for the weekend is a cinch. Photo AI Are generative images the future? The applications for AI-generated imagery are endless. It’s only a matter of time before these tools are adopted worldwide, both professionally and for everyday use. The platforms we see today are just the very beginning. Every day that goes by is more time for these AI models to learn and continue to perfect their algorithms. So whether you’re looking for an interior designer or custom images, AI holds the key to the future of generative images.

The Evolution of Generative Image Content

If you manage an online marketplace that relies on user engagements and social sharing, user-generated content and images (UGC/UGIs) are some of the most valuable assets. Review sites like Yelp let users rave about or bemoan the quality of restaurants through photos. Travel sites help proprietors attract visitors with beautiful shots and videos. Finally, photo-sharing sites like help photographers showcase their unique sensibilities and gain exposure. Unsplash However, UGIs are tricky to work with, and the penalty of getting it wrong can be steep. In this article, we’ll show you how to process UGIs programmatically, so you can build trust, improve user experience, and simplify your workflow. The Benefits of User-Generated Content User-generated content, especially images and videos, is great for building an ever-growing user base. If you build a healthy ecosystem on your site or app, you can harness the power of UGIs in these three ways: Friends, neighbors, peers, and colleagues trust each other much more than they trust marketing messages. A recent study by Stackla found that 60% of consumers based their buying decision on advice from family and friends. When the connection is authentic, there is more trust. Trust: People's tastes and preferences are unique. In a world of increased individuality, users want personalization. User-generated content is a great way to reach niche audiences and cater to their needs. Personalization: What's more, getting images from users is often the more practical and cost-effective way to generate content, especially for online marketplaces. Cost-effectiveness: What to Focus on When Processing User-Generated Images If you are working with UGC, chances are you're dealing with thousands, if not millions, of UGIs daily. The images may be from inexperienced influencers, seasoned photographers, or anyone in between. As a result, these images likely have a huge range of formats, sizes, and qualities. On the other end, your site visitors who consume the content still expect the best experience, which includes the following: Fast image loading and page speed Responsive design and device adaptability Standardized format and composition Brand safety and content moderation Bridging the gap between user expectations and the challenges of UGI processing can be daunting. And finding a simple solution that solves all these needs can seem nearly impossible. However, there are cost-effective, reliable, and scalable solutions, and in the rest of this article, we’ll show you exactly what they are and how easily you can deploy them. 1. Optimizing for Fast Loading and Page Speed Images and videos, particularly those generated by users, are often the largest files on a web page, leading to slow page speed. If SEO is important to your business, you should try to optimize and compress images, which can improve your page speed, , and in turn, your search results. With Imgix, you can combine three common techniques to reduce the image file size and speed up page speed: Core Web Vitals Serve images in next-gen formats such as AVIF or WebP Properly resize image dimensions Let Imgix apply best-effort compression In the examples below, the original image has a file size of 3.3MB. The optimized image, despite looking identical, has a file size of only 215kB, which is 93% smaller than the original. This would improve loading speed significantly. It’s achieved by using three simple transformations as an URL query string - . fm=webp&auto=compress&w=2000 Original Image - 3.3MB . Click on the link and see the size reduction. Optimized image - 215kB 2. Responsive Design With the proliferation of devices and browsers, you simply can’t rely on one image version to adapt to different environments and look good in each. Conversely, manually generating derivative files for different dimensions is labor-intensive and hardly scalable. Fortunately, the attribute alleviates that need by directing the browser to identify the best image size based on available options. Additionally, you can use Imgix and element for Art Direction. <srcset> <picture> Srcset for resolution switching <picture> for art direction 3. Standardize Visual Quality and Composition Most user-generated images aren’t created professionally. They are often too bright, too dark, or simply too inconsistent from one another. To improve the visual quality and cohesion, you should consider a standard set of image enhancements. And if you’re processing thousands or more UGIs a day, automation can help you save a massive amount of time. — makes the image more vibrant by adjusting the distribution of highlights, midtones, and shadows across RGB channels auto=true — provides red-eye removal auto=redeye Focusing and framing the faces are also important. The facial expression of a news subject, if zoomed in and cropped appropriately, can often tell a more nuanced story than any words. And if you host product reviews from users, user avatars would add authenticity and credibility. Several Imgix API parameters make it easy to detect, frame, and crop to faces: — selects a face on which to center an image. You can isolate a particular face or include multiple faces. In the example below, we selected the child’s face using and a few other parameters. faceindex faceindex=2 — defines the amount of padding around the faces. In the example below, along with a few other parameters, we zoom into the model’s face using . facepad facepad=1 — sets a mask over an image. In the example below, along with a few other parameters, we put a circular mask around the model’s face using . mask mask=ellipse Just to note, Imgix does not store face information or identify any individual in the pictures. All detection occurs on the fly and complies with EU Data Protection regulations. 4. Content Monitoring and Moderation With UGIs, content safety is another high-stake consideration. The Imgix can help you automatically detect potentially violent, adult, or otherwise questionable content. Content alerts make organizing user-generated images faster and easier, and you can automate content moderation more effectively. Asset Manager Get Started With User-Generated Image Processing As the volume, variety, and complexity of user-generated content increase, it can exacerbate the strain on you and your team. We built Imgix so you can deliver the best visual experiences to your visitors, improves page performance, and eliminate redundant work for your developers and designers. Also published here. What's shown in this article simply scratches the surface of what Imgix offers. Check out the following resources to see how Imgix can help you with your visual media needs. Learn about the common UGI transformations Create a free imgix account Read these case studies to see how our customers have worked with UGIs. Eventbrite Pexels Real Estate Wire (REW)

Dynamically Processing User-Generated Images for Trust, Personalization and Efficiency

Procesamiento dinámico de UGI para confianza, personalización y eficiencia

Traitement dynamique des UGI pour la confiance, la personnalisation et l'efficacité

विश्वास, वैयक्तिकरण और दक्षता के लिए यूजीआई को गतिशील रूप से संसाधित करना

信頼性、パーソナライゼーション、および効率性のために UGI を動的に処理する

UGIs de processamento dinâmico para confiança, personalização e eficiência

Tự động xử lý UGI để tin cậy, cá nhân hóa và hiệu quả

动态处理 UGI 以实现信任、个性化和效率

Often after one of our branding engagements, the design team at 20nine will be tasked with infusing the visual identity into their office . It usually involves taking pictures of the space, importing them into Photoshop, and designing on top of them. This helps the client visualize how their new brand will come to life in the spaces they call home. design Below are a couple of examples to illustrate how some of the deliverables usually look. As a developer at a creative agency, I try to merge technology with our current process to enhance the experience. With the rise of cheap, accessible VR headsets like Cardboard and increased smartphone processing power, I thought it would be awesome to create VR environments with our designs to really immerse the client in their potential new office. My initial idea involved just editing a regular panoramic image that would be viewable in a VR headset…but then I found Google’s Cardboard app. What’s really interesting about the Google Cardboard Camera app is that it creates 360-degree panoramic images with depth perception. It algorithmically creates a right eye image that recreates depth. It also grabs a clip of ambient audio to further add to the sensation of feeling like you’re in the picture. I was blown away by the stunning experience it created, and I wondered if I could edit the VR photos the way we do with our clients’ office spaces. At 20nine, we had been in the process of updating our own office over the last few months, so I had some internal designs already created to start testing with. I grabbed a panoramic shot of the 20nine office using the app and saved it to Google Photos so that I could download it onto my computer. The Cardboard Camera software saved it as a jpg so before doing any research, I brought it directly into Photoshop. I zoomed into one of the empty walls and overlaid one of the proposed designs on the wall. I saved it out and imported it back into the Cardboard app, but I noticed the right eye didn’t have the updated image. I jumped onto Google’s developer site to see if I could find out how the files are made up. It turns out that the VR elements are stored within the image itself. The main image is the left eye and the right eye and audio is stored as base64-encoded binary blobs in the image’s metadata. The metadata is stored using Adobe’s XMP standard. Since the base64-encoded image is large, it is written into the metadata in chunks based on the XMP specification. The end of each chunk contains a URL that looks something like “ http://ns.adobe.com/xmp/extension/590ad547fa26c83e9ddf1bc8b5e5b2bd ”. Initially I planned to use Javascript’s FileReader and some carefully crafted regular expressions to strip out the data and download the images, but merging the images back together would be an issue. So I landed on using and the command line to manage both the split and the join. Python I found the Python XMP Toolkit and it was as easy as setting the properties found in the Google Developer’s site and getting the properties using the XMP API. After grabbing the relevant base64 data, I saved the images and audio in the same folder. Below is an example of getting the right image data. ( https://python-xmp-toolkit.readthedocs.io/en/latest/ ) XMP_NS_GPHOTOS_IMAGE = u’http://ns.google.com/photos/1.0/image/' right_image_base64 = xmp.get_property(XMP_NS_GPHOTOS_IMAGE, u’GImage:Data’) Once the images were downloaded, I could overlay the same graphic onto the left and right images, save the jpg and merge everything back together using the same XMP properties. Importing it back into the app was as easy as connecting a USB to my phone and dropping the image into the apps folder. Once the image is back inside the Google Cardboard app, you can create a shareable link to send out to clients as a deliverable. This is an awesome ⏤ and surprisingly easy ⏤ way to WOW the client with something more than a static design. If you download the app, you can use the links below to check out the before and afters! Before: https://goo.gl/vrphoto/yu06arS9hHSRfnnP2 After: https://goo.gl/vrphoto/rrMrK0y54ojPwXv72

Using Virtual Reality to Brand Interiors

You've probably heard of the butterfly effect - the idea that tiny actions can have a massive impact on complex systems. That is how I look at improving company privacy policies. For twelve years, I have worked with countless startups helping them draft and edit their legal documents. With time I realized that these terms and policies could be written in a much simpler way, avoiding legal jargon and unintelligible sentences. As the General Data Protection Regulation (GDPR) came in 2018, I saw that one of its clauses supported my observations. It said that organizations must provide people with a privacy notice written “in a concise, transparent, intelligible, and easily accessible form”. In addition, it stated that visualization should be used, where appropriate. The last bit about visualizations particularly caught my attention. I spoke to my management at the time and suggested that we redesign and restructure our existing legal documents using bullet points and icons. My idea wasn’t met with enthusiasm since involving a designer meant extra time and funding. But the thought was bugging me because I knew exactly what was wrong with online legal texts and what needed to be done to improve this situation. So, I decided to set up my own company for creating better structured, shorter, simpler, and visualized legal documents. I believe that just by writing your company’s privacy policy in a clear and structured way, you can eventually influence how privacy policies are written and designed in huge organizations and mega businesses. And that, in turn, will start a wave of changes across the world. These are the three main reasons. 1. Users will be able to make smarter choices If people knew what they sign up for, I’m sure they would not put up with everything that companies are serving them. For example, you may not want to download a weather app that tracks your physical location wherever you go. Instead, you might choose to go for another weather forecast service or simply check the forecast online. Right now, the majority of privacy policies are written in a way that 99% of the people don’t have time and interest to read. Sadly, many companies are using this situation to their benefit by obscuring slippery clauses in an ocean of unintelligible text. The fewer people notice ambiguous terms, the better, right? For example, about your profile, actions online, details on who you follow, and other seemingly innocent information. But would you be OK if you knew that Instagram could read your messages and share them with their partners? you might be aware that Instagram gathers data In fact, , but they tell you this in a mind-bending way: they are doing exactly that We collect the content, communications and other information you provide when you use our Products, including when you sign up for an account, create or share content, and message or communicate with others. (..) We provide information and content to vendors and service providers who support our business, such as by providing technical infrastructure services, analyzing how our Products are used, providing customer service, facilitating payments or conducting surveys. I believe that if such information was stated openly and explicitly (e.g. We collect all the information you provide, create or share on our platform, including your direct messages with others), people would be much more cautious about what they say and do online, even in their private messages. 2. Companies will feel more accountable So, we’ve established that tech corporations feel invincible because thousands or maybe millions of users keep socializing on their platforms despite shady privacy policy clauses. Imagine a situation where more and more small and medium-sized companies across the world start writing their policies in a concise, structured, and visually appealing way. Sooner than later, even mega-companies like Google and Facebook will have to join the trend and start speaking honestly to their users. It’s not only about creating a better experience for the readers and clients. Clear and to-the-point terms and privacy policies can also save companies time and money for document revisions and translations. If the length of a legal text is reduced by 30-50% thanks to using plain language, structured lists, and icons, the translation costs will drop significantly as well. And you won’t need a legal team to make minor edits to your contracts and terms of service. Furthermore, . The GDPR made it a regulatory imperative to create privacy policies that are transparent and intelligible. easily readable legal documents are no longer a nice-to-have There are, of course, some good privacy policy examples out there, like and (see their privacy policy at the very bottom of the page). But to get the ball rolling, we need more and more businesses to join the trend of using transparent and plain language in their legal communication. BBC Juro 3. Children will be better protected The past five years have seen a dramatic increase in teenage internet users, with This figure has nearly doubled since 2015. 45% of teens saying they use the internet “almost constantly.” There is a plethora of social media and apps aimed exactly at teenagers, TikTok, Snapchat, and Instagram leading the parade. Now, look at . This document is 100 pages long and exceeds 28,000 words. That’s longer than many Master theses! How many of its teenage users (that make up ) do you think could and would read this hopeless chunk of legal text? Dividing it into sections for each region doesn’t help much, because each separate EULA looks like a whole legal catalog. TikTok’s Terms of Service 32.5% It looks like the company’s aim isn’t informing its users, but just the opposite - scaring them away from reading the terms. Let’s admit it - even us, the grownups, won’t have enough patience to dig through such a legal nightmare. And I’m telling you this as a lawyer. So, how can we expect our children to be digitally informed and protected? On the bright side, some initiatives are beginning to tackle this problem. The UK has recently issued a called "Age-appropriate design" that is binding to all online service providers operating in the UK. The code provides guidelines for communicating with children of different ages, including suggestions for audio and video prompts. Code of practice for online services The world desperately needs better legal documents I’m convinced that changing the way privacy policies are written and designed will make the world a better place. If more and more companies stop speaking “legalese” in their documents but use normal internet language instead, global corporations and even governments will in time be forced to improve their bureaucratic communication as well. And there’s no downside to this because legal documents that use plain language, short sentences, and bullet points still say all the same things and have an equal legal force. Start by giving your company’s Privacy Policy and Terms of Service a critical look. Take the longest paragraphs and try to restructure them using bullet points and shorter sentences. With time, consider hiring a designer to add visual elements to dilute the text or sign up for a tool to help you design clearer and better-looking legal documents.

Readable Legal Documents Are No Longer a "Nice-to-Have"

The cannabis business is growing at a rapid pace, and this means these businesses need a steady stream of cannabis plants, whether hemp or marijuana, to be grown in order to keep up with the supply. As cannabis is an agricultural product, an increase in demand doesn’t mean the supply can be accommodated overnight. What’s more, there are some challenges to growing cannabis plants which has necessitated the intervention of technologies such as robotics and artificial intelligence. Artificial intelligence and robots work in tandem to grow cannabis plants. The robotics are able to monitor things like moisture, nutrients, growth, and problems like mold in the physical world. Meanwhile, the AI uses the data fed to it by the robotics to make adjustments in nutrient and moisture levels in order to maintain optimal growing conditions. Companies like Farmbot, Seedo, and Agrify all produce technologies that aid in the indoor production of cannabis plants. Currently indoor growing is the best option for both hemp and marijuana plants for a number of reasons. First, cross-pollination between the two types of cannabis plants can lead to contamination - either hemp with too much THC to be legal or marijuana with not enough THC to be saleable. Second, these plants are bioaccumulators, which means that exposure to air or water pollution will remain in the plant through human consumption. Even the processing and extraction of these plants is benefiting from the use of technology. While the extraction of CBD from hemp plants is somewhat dangerous, manufacturers have found help in the form of artificial intelligence that can monitor tedious tasks that humans often tune out. This makes the process much safer in general. Artificial intelligence, robotics, and other technologies are the future of the cannabis industry. Learn more about tech for growing cannabis from the infographic below. The cannabis business is growing at a rapid pace, and this means these businesses need a steady stream of cannabis plants, whether hemp or marijuana, to be grown in order to keep up with the supply. As cannabis is an agricultural product, an increase in demand doesn’t mean the supply can be accommodated overnight. What’s more, there are some challenges to growing cannabis plants which has necessitated the intervention of technologies such as robotics and artificial intelligence. Artificial intelligence and robots work in tandem to grow cannabis plants. The robotics are able to monitor things like moisture, nutrients, growth, and problems like mold in the physical world. Meanwhile, the AI uses the data fed to it by the robotics to make adjustments in nutrient and moisture levels in order to maintain optimal growing conditions. Companies like Farmbot, Seedo, and Agrify all produce technologies that aid in the indoor production of cannabis plants. Currently indoor growing is the best option for both hemp and marijuana plants for a number of reasons. First, cross-pollination between the two types of cannabis plants can lead to contamination - either hemp with too much THC to be legal or marijuana with not enough THC to be saleable. Second, these plants are bioaccumulators, which means that exposure to air or water pollution will remain in the plant through human consumption. Even the processing and extraction of these plants is benefiting from the use of technology. While the extraction of CBD from hemp plants is somewhat dangerous, manufacturers have found help in the form of artificial intelligence that can monitor tedious tasks that humans often tune out. This makes the process much safer in general. Artificial intelligence, robotics, and other technologies are the future of the cannabis industry. Learn more about tech for growing cannabis from the infographic below. tech for growing cannabis

Stories, News & Mentions about UGI

Ugi’s Stories on HackerNoon

Ugi's News & Mentions