Research & Design
Are you looking to develop a new online service or software product? Do you have any of the following questions?
- I am not sure what is better for our new product; to develop it using licensed software or open source?
- I am not sure what technologies to use for the service I want built? How will my decision affect the maintenance of the service in the long run?
- I want to create an iPhone app but I am not sure if it can be done? (ex: Does Apple allow for manipulation of?)
- Will force.com (salesforce.com’s platform) benefit the application I want to build? How will it affect my operations costs over time?
- We are thinking of a new Blackberry app, but eventually we want to build the same app for the iphone and for Android. What should we be concerned about now if the app will run on multiple platforms in the future?
- I have an idea for a new application, but I need something to show to potential investors and to use to validate the app with potential customers, what is best to get started with quickly?
If you have these types of questions or similar questions Softjourn provides architecture and design services, answering those technical questions that you need to have answered. When you are to the point of needing to have something to show potential investors and/or potential clients, Softjourn works with you to determine if a wide or deep prototype is the best fit for your company and we also can build it for you.
Contact us to learn more about Softjourn's services. You can also read below the details of several of Softjourn's Research and Design projects.
Mobile Calendaring app – Blackberry/iPhone/Android research and design/Prototype design:
|
Issue: |
An entrepreneur approached Softjourn with an idea for a new mobile calendaring application. Initially it was to be rolled out for the Blackberry platform, and later to other platforms. But he needed assistance in several areas:
|
| Solution: |
Initially our entrepreneur wanted to design and develop his new mobile app only for the Blackberry platform. In this case discussion and design was revolving around the use of Blackberry push services for messenging. However, as discussions continued, Softjourn suggested that since our entrepreneur eventually wanted the app to work across multiple platforms, it was best to design for that scenario from the very beginning. Unfortunately not all mobile platforms support Push Services. For some platforms these services do not even exist, or if they exist, they are different. In this case, the middleware server cannot be the same for different platforms, it would require different logic for each platform.
A better solution was to develop a common middleware server with an interchangeable interface that could be used for each platform identically. In this scenario each client would be responsible for data retrieval from the server, via SOAP. The disadvantages of this variant include: i. possible increase in the server load (needs more powerful servers), ii. Impact on the device's battery life (i.e. launching the application taking longer and draining the battery more). The advantage of this variant is of course that less code needs to be developed and maintained for the server side.
|
| Result: |
Our entrepreneur had a better idea of his product, had the design of the server side that would work across all platforms, and the design of his proof of concept for showing his potential clients and investors.
“You have asked the best questions and are very thorough!”
Entrepreneur & Business Owner
|
Check 21 application: Analyze and select OCR recognition software for a money transfer service:
|
Issue: |
Softjourn was selected to develop a new Check21 application. As part of that project, scanners were selected by our client for scanning the checks; however, the RDM scanners chosen at that time included reading the MICR line and did two sided scanning, but they did not include OCR software to recognize the check amount. Therefore Softjourn needed to evaluate and select appropriate OCR software keeping in mind the volume of transactions expected to be processed at the initial release of the new Check21 application and in the future, as well as the budget our client had for purchasing OCR software.
|
| Solution: |
Initially Softjourn created a long list of potential OCR vendors; including open source vendors and large-scale commercial products. The Check21 application being developed for our client was to replace the one that our client paid for on a subscription basis, from USBank. USBank’s solution included high-end OCR recognition including recognizing CAR – Courtesy Amount Recognition (the amount in characters) and LAR – Legal Amount Recognition (the hand written amount). The solution Softjourn chose also had to take in to account that almost 50% of the checks processed by the new system were going to be hand written, so the OCR software would have to be able to recognize the amounts on a lot of different checks. Having a .NET version was also a requirement as the new Check21 system was being written in .NET.
Some of the vendors evaluated included:
As well as several higher-end solutions:
Evaluation included installing all of the different products, creating test image files based on real check images, supplied by our client, and comparing the results.
|
| Result: |
After initial results and creating a short list of possible vendors, Softjourn recommended the QuickStrokes Banking Edition to its client. The product included ICR, OCR, MICR, CAR and Business LAR. It did not include hand-printed LAR and was therefore less expensive. Mitek did have a QuickStrokes Premier Banking Edition which could read the LAR field, but in most cases it was not necessary. Recognition results averaged 92% with the Premier edition and 78% with the regular Banking edition. In the end our client agreed that based on the expected recognition results and the pricing difference, the Banking edition was the right choice for them.
|
OCR recognition software research for a digital marketing company:
|
Issue: |
Softjourn was approached by a digital marketing company that wanted to explore an idea to use a mobile app to read key cards which are part of an identification system and would be used to access websites, online banking services, etc. One of the requirements for this application was that it should be written using open source libraries.
|
| Research: |
Before starting the actual development Softjourn suggested to first conduct research, involving analyzing a number of free Optical Character Recognition (OCR) libraries, in order to determine which one would provide the best results.
The Softjourn team analyzed a number of free OCR SDKs (Software Development Kit’s for adding Optical Character Recognition), eight to be exact.
For comparison between commercial and freeware OCR SDKs, Softjourn compared the results using ABBYY FineReader.
|
| Result: |
Not surprisingly, the tests showed that the best recognition results were provided by the commercial OCR product (ABBYY FineReader).
Out of the freeware OCR SDKs evaluated, Tesseract showed the best results. The other free libraries were not a good fit for different reasons, e.g. GPL license did not allow selling applications; no source code available for iPhone, freeware supported only Windows OS, etc.
|
| Technologies: |
OCR SDKs:
|
Automating the process of reviewing audio and video files posted to websites:
|
Issue: |
A marketing firm was interested in what their options were for automating the process of reviewing audio and videos posted to their online service. For example if there was greater than x% of nudity the video should be flagged and not posted. Or if a video contained bad language it should not be posted.
|
| Research: |
There are a number of tools available to “automate” the process of determining whether or not a written post should be deleted or flagged as unacceptable.
Automated censorship of videos can be divided in two tasks:
Automated censorship of video stream in its own turn could be reduced to the censorship of individual frames (with possible optimizations regarding human perception) thus it is the same as automated censorship of photos.
Audio stream area of monitoring could be bad (swearing) language.
Photos (video frames) area of monitoring could be for nudity, swear (bad) language drawings/prints, violence and other inappropriate material.
Audio Stream Monitoring
There are patented solutions for an input audio data stream comprising speech processing by an automatic censoring filter in either a real-time mode, or a batch mode, producing censored speech that has been altered so that undesired words or phrases are either unintelligible or inaudible. For example US patent 7437290 assigned to Microsoft corporation.
Photos (Video Frames) Monitoring could be divided in a sub-set of tasks as:
Detection for each of the different types of photo (video frames) monitoring can be further divided, for example bad language detection can be divided in to:
The main issue here can be false-positive marks on parts of the words or foreign words. The solution to this issue is so-called “Language Guesser” systems based on categorization algorithm presented in Cavnar, W. B. and J. M. Trenkle, “N-Gram-Based Text Categorization''.
Violence detection also has its own specifics.
|
| Result: | As part of this project Softjourn also evaluated a number of existing software products which could be used for nudity detection, and made recommendations as to which product suited our customer’s online service and what type of customization could be made to address, to the extent possible the other types of automated detection necessary. |
| Technologies: |
Products evaluated:
|