Video Analytics Case Studies

Part
01
of one
Part
01

Video Analytics Case Studies

After an extensive search through corporate reports, news articles and press releases based in both the US and China, and case studies regarding video surveillance, case studies of companies within the convenience store space that use video analytics in their existing video camera systems do not appear to be available in the public domain. However, the research team was able to gather valuable insights about stores and companies that already have camera set-ups and use that feed for analytics.


USEFUL FINDINGS:

CASE #1: SPRINT FOOD STORES, USA
  • Downsides to its use:
    • None of the publicly available information indicated any downsides of the use of video analytics at Sprint Food Stores specifically.
    • However, according to the ACLU's report on the use of this technology in stores, this carries the risk of being misidentified as someone in a database of suspected criminals, fugitives, terrorists, or whatever other blacklists stores may be using or begin using in the future.
    • Also, this may encroach upon the privacy of the customers as they do not expect that each of their movements within the store to be analyzed by computers and filed away to give insight into their shopping habits, patterns, and preferences.

CASE #2: BROOKSTONE STORES, USA
  • This example is from the retail store chain Brookstone US.
  • It is important to note that Brookstone had 307 stores, including 48 airport-based stores and 3 outlet stores in 42 states and Puerto Rico. However, it went out of business on April 1, 2019.
  • However, the company's example has been included in our findings because of its reuse of an existing video camera system for analytics purposes and had been using it since 2013.
  • How they used video analytics:
    • Brookstone used it to determine the "days and times with the highest customer traffic to optimize staffing."
    • According to the National Retail Federation (NRF), Brookstone "also used analytics to determine how long customers stayed in one area helping to place employees and products in optimal places in the store."
    • The NRF also notes that "merchandisers can use the same technology to determine which displays attract more traffic and marketers can measure promotions.
  • How it benefited them:
    • None of the publicly available information indicated any downsides of the use of video analytics in Brookstone stores specifically.
    • However, according to the New York Times, since "video analytics software is often trained on publicly available footage, such as YouTube videos," it may have a bias or be entirely inaccurate due to the "people who post them or in what such videos show." Therefore, there is a chance of misjudgment when using video surveillance analytics.
CASE #3: JACKSON FOOD STORES, USA
  • The following example relates to the use of video analytics in Jackson Food Stores, which has nearly 245 stores across the US.
  • How they used video analytics:
    • Before customers can enter the basic convenience store, a camera under a red awning is supposed to take a picture and then analyze it to determine whether the image matched with any in a database of known robbers and shoplifters at that location.
  • How it benefits them:
    • If recognized, the security platform takes the limited action of locking a door on suspected shoplifters.
    • The companies using similar technology have reported a 95% reduction in police service calls, while shoplifting theft was cut in half.

RESEARCH STRATEGY:

To find examples of companies within the convenience store space that use video analytics in their video cameras, the research team deployed the following strategies:

REPUTED NEWS SOURCES FROM THE SPECIFIED REGIONS
To find case studies of companies within the convenience store space that use video analytics in their video cameras, the research team looked into reputable news sources in both China and the US. Although we found some examples of retail outlets like Freshhippo and 7fresh, which use technologies including video analytics in their stores, the examples were not related to the reuse of their current set-up. The objective of the use of video analytics technology in these stores was to freshly deploy new technologies to make them suitable for self-checkout or operating unmanned. We then also looked for examples pertaining to the US, searching news publications that discuss examples of big stores such as Walmart, Kroger, and Target, but the examples, similar to our findings for China, were focused upon the installation of an entirely new video feed set up. Therefore, the examples could not be considered.

TECH SOURCES
Since the scope of the request deals with the deployment of technology, the research team looked into sources that cover exclusive technological content. Our aim was to find out details relating to video analytics technology and then to look into the use cases of those technologies in China and the US. There were two kinds of examples available. The first kind relates exclusively to the features and capabilities of surveillance cameras and the quality of the videos they produce. The second kind regarded the use of various new technologies including video analytics, leading toward an Amazon Go model. Given that we were looking for case studies that were dissimilar to the Amazon Go model, the research team considered sources that talked about video surveillance. We then looked into whether any of the stores that were using video surveillance integrated it with an analytics utility or not. Given this search, we were unable to find any convenience stores utilizing this model in either the US or China.

INTERVIEWS WITH COMPANY EXECUTIVES AND COMPANY REPORTS
We then located the C-level executives of companies like Walmart, Alibaba, Target, and Kroger. We used this information to search for interviews with these executives in reputable publications based in both the US and China. Our objective was to find out whether they mentioned any such incidence where a current video camera set up was reused. It was found that in China, Alibaba, through its flagship initiative Alizila, was in the process of overhauling millions of Mom-and-Pop convenience stores into modern-day stores. Considering that most of these old stores hardly had any existing video-surveillance facility, there was no information on the current set-up being reused for video analysis. Similarly, Kroger, which plans to come up with several self-checkout stores across the US in the coming years, is mostly concerned with freshly deployed technology.

CASE STUDIES RELATING TO VIDEO SURVEILLANCE AND ITS USE IN ANALYTICS
For this strategy, we looked into sources such as reports from both academic institutions and national and international organizations regarding the trends of video surveillance. Our objective was to find out whether these surveillance efforts led to analytics in China or the US. Although there were reports covering the pros and cons of surveillance, there were hardly any specific instances of case studies pertaining to convenience stores.

CONCLUSION
In view of the above limitations and challenges, the research team provided the three cases in the findings above:
  • Sprint Food Stores, USA, and Jackson Food Stores, USA: They were using their video footage for further analysis. Although it was not clearly ascertainable whether it was a reuse of current set up for analysis.
  • Brookstone Stores USA: This was included because the company sells many products like those sold in a convenience store, such as food and wine, personal care, and home use products. However, Brookstone stores went out of operation in April 2019. However, since it was using analysis of its existing surveillance videos since 2013, it was included as a useful finding.
Sources
Sources

Quotes
  • "The cameras provide easily retrievable and emailable surveillance video evidence (with clearer images) that help deter theft and fraud and enable the retailer to efficiently monitor store conditions from home, headquarters, or anywhere with an Internet connection."
  • "With an advanced surveillance camera system, a convenience store chain could achieve ROI within a year through better prevention of theft, fraud, and unjustified claims as well as improved operational oversight,” said Todd Harrison, IT director, Sprint Food Stores, which operates 20 convenience stores in Georgia and South Carolina. Harrison also oversees loss prevention camera surveillance for the company."
  • "The surveillance system can also help to identify and deter potential cash register transaction theft. It integrates with the store POS system and provides smart ER exception reports, a type of advanced filtering software included with the surveillance system. It allows for quick analysis of thousands of register transactions to identify suspicious activity and pull up video of it for review."
  • "Video can be reviewed and archived remotely as needed, which eliminates the need to spend hours every week traveling to burn DVDs for video storage,” says Harrison."
  • "The video analytics module would basically pinpoint where in the store people are traveling, where they stop, and where they spend time, over time,” said Harrison."
  • "When a store was involved with a slip and fall claim, we were able to show on video that other customers took the same path as the accuser without incident, and that less harm was involved than stated,” says Harrison. “We retrieved the video, sent it to her lawyer, and her lawyer dropped the case.” Surveillance video can also be used as evidence to collect costly damage to equipment or store property. This helped when a truck driver ran into a 3-foot high, LED diesel fuel sign, destroying it, at a Sprint location before driving off"
Quotes
  • "Where once video was used in VHS form for surveillance, retailers are now “using video analytics to capture customer trends, track traffic, catch shoplifters and reduce fraud,” according to a blog posted by Jennifer Overstreet of the National Retail Federation (NRF), a global retail trade association with members representing over 3.5 million establishments in the U.S."
  • "Brookstone and American Apparel (speakers at the NRF’s 2013 annual conference) use it to determine days and times with the highest customer traffic to optimize staffing,” according to the NRF blog. “Analyzing “dwell or how long customers stay in one area helps place employees and products in optimal places in the store. Merchandisers can use the same technology to determine which displays attract more traffic and marketers can measure promotions."
  • "Both retailers saw conversions (to sales) increase and losses (from shoplifting etc.) decrease, according to the posting."
Quotes
  • "Before patrons can enter the basic convenience store at the corner of South 38th Street and Pacific Avenue, a camera under a red awning will take a picture and use artificial intelligence (AI) to decide whether the image matches any in a database of known robbers and shoplifters at that location."
  • "That's a privacy violation because you should be notified about it," Diharce said on a recent morning. "They should have a sign to notify you that they're comparing it to photos of criminals."
  • "A customer named May, who asked that her surname not be used, said she had no issue with the surveillance tool, but questioned its purpose in a corner store. "It's a little store – I don't see why they need something like that," May said as she stood in the Jacksons parking lot."
  • "Blue Line Technology senior partner Tom Sawyer said the more than 20 retail companies that use its software platform throughout the nation have reported a 95% reduction in police service calls, and shoplifting theft is cut in half."
  • "Blue Line Technology's security platform takes the limited action of locking a door on suspected shoplifters"
Quotes
  • "But I think it’s fair to say that most customers do not think that they are being subject to a perpetual lineup, scrutinized by face recognition technology to see if they resemble anyone that a company security service has decided to put on a watch list. They do not expect that their faces are being captured, retained, connected to their real-world identity (for example when they use a credit card at checkout), and combined with information about their income, education, demographics, and other data. "
  • "They do not expect that their every footstep, hand motion, and gaze will be analyzed by computers and filed away to give insight into their shopping habits, patterns, and preferences, and shared among different companies, data brokers, and advertisers. They do not expect that they are subject to the risk of being misidentified as someone in a database of suspected criminals, fugitives, terrorists, or whatever other blacklists stores may be using or begin using in the future. They don’t expect that all these intimate details about their behavior will become accessible to government agencies through legal demands or voluntary sharing."
  • "And if those things are happening, I think most customers would want to know about it."
Quotes
  • "Video analytics software is often trained on publicly available footage, such as YouTube videos, but there may be bias in the kinds of people who post them or in what such videos show."
  • "The use of video analytics may also have a chilling effect on society, the A.C.L.U. warned. If individuals feel that their every movement is monitored, they may alter their behavior. It may also lead to over-enforcement of smaller crimes, a practice that has disproportionately affected minorities or other disadvantaged groups."