Find Jobs
Hire Freelancers

Visual Identification & Segmentation of Scanned Documents in PDF

min $50 USD / hour

Imefungwa
Imechapishwa over 5 years ago

min $50 USD / hour

I have a PDF file filled with scanned documents. These are standardized form documents that have IDs, standard formats, and distinguishable segments (although certain sections of a form may expand or contract in length in each instance). Also included in this PDF collection of scanned pages are identification documents (driver’s license, passport, etc). These are more sensitive documents and should be closely scrutinized from a privacy/security perspective. I am looking for someone to help design the appropriate image recognition capabilities to train a machine how to associate the many segments across the many documents in each PDF and then classify them as “evidence” of categorical questions (classifications with confidence). Additionally, as this “evidence” is classified for each PDF, I need the system to capture a “snapshot” of that particular segment for quick visual reference and confirmation by humans. Humans need the ability to adjust the segment “snapshots” identified by the system and the ability to adjust the classification of each segment. I would like to vegan with a proof-of-concept that will price the viability of the conceived AI system. Once proven, further investment will be made to scale and, eventually, automate recommendations/actions for humans using an RPA (Robitic Process Automation) system. Link to potentially helpful reference: [login to view URL]
Kitambulisho cha mradi: 17396823

Kuhusu mradi

20 mapendekezo
Mradi wa mbali
Inatumika 6 yrs ago

Unatafuta kupata pesa?

Faida za kutoa zabuni kwenye Freelancer

Weka bajeti yako na muda uliopangwa
Pata malipo kwa kazi yako
Eleza pendekezo lako
Ni bure kujiandikisha na kutoa zabuni kwa kazi
20 wafanyakazi huru wana zabuni kwa wastani $54 USD/saa kwa kazi hii
Picha ya Mtumiaji
Hi there! Thanks for posting the job. I am skilled and experienced in image processing and OCR programming so I can help you with this project. I hope to discuss the project further with you. Thank you!
$58 USD ndani ya siku 40
5.0 (8 hakiki)
6.5
6.5
Picha ya Mtumiaji
Hi there, We are a team based in Argentina with strong experience in software and mobile development. We are able to do your project starting ASAP. Do not hesitate to contact me if you are interested. You will not regret, Thank you very much, Best regards, Jorge.
$58 USD ndani ya siku 40
5.0 (1 hakiki)
5.5
5.5
Picha ya Mtumiaji
Hello, I analyzed the project description and I understand that is required to develop a document understanding and classification solution. The documents are extracted from PDF pages. Further it will be required to develop a machine learning application to train the system with a training set to be used later as system to detect and classify the documents. This can be a challenging project depending on the required accuracy. Of course as the first step needs to be made always a proof of concept. I am available to discuss the details and my plan of action and development if you are interested. Best regards, Patrik
$58 USD ndani ya siku 40
4.8 (7 hakiki)
5.4
5.4
Picha ya Mtumiaji
hello I am very interested in this project. I am a computer vision expert. I have good experiences in this kind of project. Please check my portfolio. I can do this job perfectly. Thank you.
$55 USD ndani ya siku 40
5.0 (2 hakiki)
3.4
3.4
Picha ya Mtumiaji
How are you ? My name is Bryan Valencia. Please check my profile. https://www.freelancer.com/u/UnitedWebSpace Looking forward to discuss about your project. Thank you. Bryan.
$60 USD ndani ya siku 40
3.8 (1 hakiki)
2.7
2.7
Picha ya Mtumiaji
Hi, I am an engineer with years of experience in machine learning. I work mostly on vision based research and development projects using deep learning. I have a very good knowledge in various machine learning and computer vision algorithms. Some of the projects I have worked are face recognition, attribute detection (age, gender, etc), people re-identification, object detection, semantic segmentation using deep learning. I clearly understand the task described by you, and I have several approaches in mind on how this can be done. However I might be necessary to train a model for detection+classification of the segment in a page, therefore sufficient training/testing data might be required. I would like to discuss more on this, since this sounds to be an interesting project. This my knowledge and experience, I will be able to successfully complete this task. Regards, Dilina.
$50 USD ndani ya siku 40
5.0 (1 hakiki)
2.2
2.2
Picha ya Mtumiaji
We are team of Scientists and Developers having rich experience with Artificial Intelligence Techniques and Machine Learning Techniques like Neural and NLP. We have done extensive research on Facial Recognition and Computer Vision & Image Processing. The Technical team consists of Programmers having experience more than 6 years.
$50 USD ndani ya siku 3
5.0 (2 hakiki)
1.3
1.3
Picha ya Mtumiaji
Hi, We can help you create a site that leverages responsive design. Our designers and developers will work together to create an attractive site that uses the latest web technologies and delivers a superior experience to visitors. We can redesign your entire site, make minor adjustments, or provide our expert advice. My name is Bill, and I am the Business Analyst at GOLPIK. We are a Web Design & Development company, based in Lombard, IL with a phenomenally talented team of designers and developers. We appreciate the opportunity to work with you. Regards, Bill Golpik
$55 USD ndani ya siku 40
5.0 (1 hakiki)
1.4
1.4
Picha ya Mtumiaji
Hi I am Hitesh. I have a team of experienced Machine Learning, Deep Learning, AI and Python developers with great hands-on experience. We have already worked on facial recognition system and trained model based on various techniques of Machine Learning. Let's discuss further on this project. Thanks Hitesh
$50 USD ndani ya siku 40
5.0 (1 hakiki)
1.0
1.0
Picha ya Mtumiaji
Hi, I can see that you want to do some revolutionary thing here, congratulation for that. Now if we talk about the project, I can understand that you want an MVP of Image Recognition System that has the capability to capture a snapshot of a particular segment for quick visual reference which leads to close scrutinization of sensitive documents. I have read the attached Research paper & tried to understand PLRS vs LEX Approach and I think I can help you with it. Please initiate the chat so that can discuss in detail. I look forward to hearing from you & working on this interesting project. With regards, Ezaj A
$55 USD ndani ya siku 40
5.0 (1 hakiki)
0.5
0.5
Picha ya Mtumiaji
Over my career I have developed a wide range of applications and websites from desktop and mobile applications to enterprise servers, embedded systems and responsive websites, through to to video, and static sites and have significant experience with PHP having written tens of thousands of lines of PHP. I have architected databases, written frameworks and and other solutions ranging from trading applications to dynamic websites and business management tools. I have developed both responsive websites and server side responsive solutions and written pagination infrastructure, and Ajax back ends. I have also managed teams, from being an individual contributors to large teams, both onshore and offshore. I have a Computer Science / Micro-Electronics degree and have worked on many system integrations, APIs and integration modules and visual components and can recoginize appropriate integration solutions for your site. I have experience with mysql, Linux, HTML, JavaScript, Ajax, CSS, CDNs, AWS, git, github and other tools over many years and can bring a wide range of known solutions to bare on a problem. I have specific experience with many aspects of financial services, payments solutions, micro-electronics and the publishing industry. I am able to be flexible on the work hours and arrangements and can operate with a control on the maximum hours per month if necessary. I am based in New York City, so am in the Easter time-zone. Regards, Owen.
$55 USD ndani ya siku 40
0.0 (0 hakiki)
0.0
0.0
Picha ya Mtumiaji
We offer end to end data curation and enrichment services with an ability to work on both Enterprise platforms as well as capability to ingest clients’ data into proprietary in-house platform, ensuring Quality, Efficiency and Model Maturity. Core focus of our software development services is on Emerging Technologies, working on innovative solutions and custom software development for our esteemed customers. AI Touch core team, having worked on cutting edge industry use cases of Computer Vision, Text and Image classification, realize that AI models take data and time to mature. AI Touch provides: 1. Data Enrichment Services: Data Labeling / Annotations; AI / ML Model Tuning and Testing; Live Operations Support 2. Software Development : Innovation solutions for Emerging Technologies (Artificial Intelligence, Machine Learning, Deep Learning, Internet of Things, Augmented Reality / Virtual Reality / Mixed Reality, NLP, Computer Vision etc); Custom Software Development and QA/Security Assessment Our In-house Proprietary Platforms are: 1. Data Labeling Toolkit: One stop solution for all types of annotation requirements (Image/Text/Video); comprehensive analytics and reporting; smart integrations with Enterprise AI/ML Models for model training and testing 2. Employee Monitoring System: Team Productivity and Efficiency monitoring; integrable with Enterprise BI Systems to track and apply continuous improvement BPM measures
$55 USD ndani ya siku 40
0.0 (0 hakiki)
0.0
0.0
Picha ya Mtumiaji
Hi there, I have been working on border control systems, dealing with personal identification document parsing and sensitive data handling. In this context, I have developed a system based on template-based content extraction for passports, that enables to obtain the multiple section in the passport main page and their respective text content. For the current project, I propose to develop a system composed of the following steps: 1. Classification of the document type to identify the different documents in the pdf; 2. Text segments detection on each document for improving subsequent processing steps; 3. Parsing the document content to obtain the text in each segment; 4. Association models that will allow evidence association across documents. This hierarchical approach will allow to build a more reliable system, since each step will have less variability. For example, having a text detection and OCR step for all the document types will most likely fail because of all the possible document layout variations. Milestones: 1. Data analysis, annotation and region segmentation to build the detection models - 32 hours (depending on annotation complexity) 2. Document classification model - 24 hours 3. Text detection model - 40 hours 4. OCR text parsing - 16 hours 5. Association models and snapshot system - 40 hours (depending on the complexity of the text search method, and type of association required). Best regards, Vitor Pedro
$50 USD ndani ya siku 40
0.0 (0 hakiki)
0.0
0.0
Picha ya Mtumiaji
Hi , We have Senoir developer in our company .We are interested in this assignment . Thanks Santosh
$55 USD ndani ya siku 40
0.0 (0 hakiki)
0.0
0.0
Picha ya Mtumiaji
Hey there i can robotize your task in a minimal time Feel free to contact waiting for your response Thanks
$55 USD ndani ya siku 40
0.0 (0 hakiki)
0.0
0.0
Picha ya Mtumiaji
I have finished the official Advanced Training from UI Path, and I would love to accomplish your tasks at hand. Thank your for your time.
$57 USD ndani ya siku 20
0.0 (0 hakiki)
0.0
0.0
Picha ya Mtumiaji
We have experience working with AWS image recognition platforms, and have also been developing on their new DeepLens platform. We believe we can leverage our experience in both areas (training with edge data as well as storage buckets) and develop a solution for your business application! Hope to hear from you so that we can further evaluate this project!
$52 USD ndani ya siku 40
0.0 (0 hakiki)
0.0
0.0

Kuhusu mteja

Bedera ya UNITED STATES
United States
0.0
0
Njia ya malipo imethibitishwa
Mwanachama tangu Jul 19, 2018

Uthibitishaji wa Mteja

Asante! Tumekutumia kiungo cha kudai mkopo wako bila malipo kwa barua pepe.
Hitilafu fulani imetokea wakati wa kutuma barua pepe yako. Tafadhali jaribu tena.
Watumiaji Waliosajiliwa Jumla ya Kazi Zilizochapishwa
Freelancer ® is a registered Trademark of Freelancer Technology Pty Limited (ACN 142 189 759)
Copyright © 2024 Freelancer Technology Pty Limited (ACN 142 189 759)
Onyesho la kukagua linapakia
Ruhusa imetolewa kwa Uwekaji wa Kijiografia.
Muda wako wa kuingia umeisha na umetoka nje. Tafadhali ingia tena.