Recherche avancée

Sur d’autres sites (6375)

• Developing MobyCAIRO

  26 mai 2021, par Multimedia Mike — General

  I recently published a tool called MobyCAIRO. The ‘CAIRO’ part stands for Computer-Assisted Image ROtation, while the ‘Moby’ prefix refers to its role in helping process artifact image scans to submit to the MobyGames database. The tool is meant to provide an accelerated workflow for rotating and cropping image scans. It works on both Windows and Linux. Hopefully, it can solve similar workflow problems for other people.

  As of this writing, MobyCAIRO has not been tested on Mac OS X yet– I expect some issues there that should be easily solvable if someone cares to test it.

  The rest of this post describes my motivations and how I arrived at the solution.

  Background
  I have scanned well in excess of 2100 images for MobyGames and other purposes in the past 16 years or so. The workflow looks like this :

  
  

  

  Image workflow

  
  

  It should be noted that my original workflow featured me manually rotating the artifact on the scanner bed in order to ensure straightness, because I guess I thought that rotate functions in image editing programs constituted dark, unholy magic or something. So my workflow used to be even more arduous :

  
  

  

  I can’t believe I had the patience to do this for hundreds of scans

  
  

  Sometime last year, I was sitting down to perform some more scanning and found myself dreading the oncoming tedium of straightening and cropping the images. This prompted a pivotal question :

  
  Why can’t a computer do this for me ?
  

  After all, I have always been a huge proponent of making computers handle the most tedious, repetitive, mind-numbing, and error-prone tasks. So I did some web searching to find if there were any solutions that dealt with this. I also consulted with some like-minded folks who have to cope with the same tedious workflow.

  I came up empty-handed. So I endeavored to develop my own solution.

  Problem Statement and Prior Work
  
  I want to develop a workflow that can automatically rotate an image so that it is straight, and also find the most likely crop rectangle, uniformly whitening the area outside of the crop area (in the case of circles).

  As mentioned, I checked to see if any other programs can handle this, starting with my usual workhorse, Photoshop Elements. But I can’t expect the trimmed down version to do everything. I tried to find out if its big brother could handle the task, but couldn’t find a definitive answer on that. Nor could I find any other tools that seem to take an interest in optimizing this particular workflow.

  When I brought this up to some peers, I received some suggestions, including an idea that the venerable GIMP had a feature like this, but I could not find any evidence. Further, I would get responses of “Program XYZ can do image rotation and cropping.” I had to tamp down on the snark to avoid saying “Wow ! An image editor that can perform rotation AND cropping ? What a game-changer !” Rotation and cropping features are table stakes for any halfway competent image editor for the last 25 or so years at least. I am hoping to find or create a program which can lend a bit of programmatic assistance to the task.

  Why can’t other programs handle this ? The answer seems fairly obvious : Image editing tools are general tools and I want a highly customized workflow. It’s not reasonable to expect a turnkey solution to do this.

  Brainstorming An Approach
  I started with the happiest of happy cases— A disc that needed archiving (a marketing/press assets CD-ROM from a video game company, contents described here) which appeared to have some pretty clear straight lines :

  
  
  

  My idea was to try to find straight lines in the image and then rotate the image so that the image is parallel to the horizontal based on the longest single straight line detected.

  I just needed to figure out how to find a straight line inside of an image. Fortunately, I quickly learned that this is very much a solved problem thanks to something called the Hough transform. As a bonus, I read that this is also the tool I would want to use for finding circles, when I got to that part. The nice thing about knowing the formal algorithm to use is being able to find efficient, optimized libraries which already implement it.

  Early Prototype
  A little searching for how to perform a Hough transform in Python led me first to scikit. I was able to rapidly produce a prototype that did some basic image processing. However, running the Hough transform directly on the image and rotating according to the longest line segment discovered turned out not to yield expected results.

  
  
  

  It also took a very long time to chew on the 3300×3300 raw image– certainly longer than I care to wait for an accelerated workflow concept. The key, however, is that you are apparently not supposed to run the Hough transform on a raw image– you need to compute the edges first, and then attempt to determine which edges are ‘straight’. The recommended algorithm for this step is the Canny edge detector. After applying this, I get the expected rotation :

  
  
  

  The algorithm also completes in a few seconds. So this is a good early result and I was feeling pretty confident. But, again– happiest of happy cases. I should also mention at this point that I had originally envisioned a tool that I would simply run against a scanned image and it would automatically/magically make the image straight, followed by a perfect crop.

  Along came my MobyGames comrade Foxhack to disabuse me of the hope of ever developing a fully automated tool. Just try and find a usefully long straight line in this :

  
  

  Darn it, Foxhack…
  

  There are straight edges, to be sure. But my initial brainstorm of rotating according to the longest straight edge looks infeasible. Further, it’s at this point that we start brainstorming that perhaps we could match on ratings badges such as the standard ESRB badges omnipresent on U.S. video games. This gets into feature detection and complicates things.

  This Needs To Be Interactive
  At this point in the effort, I came to terms with the fact that the solution will need to have some element of interactivity. I will also need to get out of my safe Linux haven and figure out how to develop this on a Windows desktop, something I am not experienced with.

  I initially dreamed up an impressive beast of a program written in C++ that leverages Windows desktop GUI frameworks, OpenGL for display and real-time rotation, GPU acceleration for image analysis and processing tricks, and some novel input concepts. I thought GPU acceleration would be crucial since I have a fairly good GPU on my main Windows desktop and I hear that these things are pretty good at image processing.

  I created a list of prototyping tasks on a Trello board and made a decent amount of headway on prototyping all the various pieces that I would need to tie together in order to make this a reality. But it was ultimately slowgoing when you can only grab an hour or 2 here and there to try to get anything done.

  Settling On A Solution
  Recently, I was determined to get a set of old shareware discs archived. I ripped the data a year ago but I was blocked on the scanning task because I knew that would also involve tedious straightening and cropping. So I finally got all the scans done, which was reasonably quick. But I was determined to not manually post-process them.

  This was fairly recent, but I can’t quite recall how I managed to come across the OpenCV library and its Python bindings. OpenCV is an amazing library that provides a significant toolbox for performing image processing tasks. Not only that, it provides “just enough” UI primitives to be able to quickly create a basic GUI for your program, including image display via multiple windows, buttons, and keyboard/mouse input. Furthermore, OpenCV seems to be plenty fast enough to do everything I need in real time, just with (accelerated where appropriate) CPU processing.

  So I went to work porting the ideas from the simple standalone Python/scikit tool. I thought of a refinement to the straight line detector– instead of just finding the longest straight edge, it creates a histogram of 360 rotation angles, and builds a list of lines corresponding to each angle. Then it sorts the angles by cumulative line length and allows the user to iterate through this list, which will hopefully provide the most likely straightened angle up front. Further, the tool allows making fine adjustments by 1/10 of an angle via the keyboard, not the mouse. It does all this while highlighting in red the straight line segments that are parallel to the horizontal axis, per the current candidate angle.

  
  
  

  The tool draws a light-colored grid over the frame to aid the user in visually verifying the straightness of the image. Further, the program has a mode that allows the user to see the algorithm’s detected edges :

  
  
  

  For the cropping phase, the program uses the Hough circle transform in a similar manner, finding the most likely circles (if the image to be processed is supposed to be a circle) and allowing the user to cycle among them while making precise adjustments via the keyboard, again, rather than the mouse.

  
  
  

  Running the Hough circle transform is a significantly more intensive operation than the line transform. When I ran it on a full 3300×3300 image, it ran for a long time. I didn’t let it run longer than a minute before forcibly ending the program. Is this approach unworkable ? Not quite– It turns out that the transform is just as effective when shrinking the image to 400×400, and completes in under 2 seconds on my Core i5 CPU.

  For rectangular cropping, I just settled on using OpenCV’s built-in region-of-interest (ROI) facility. I tried to intelligently find the best candidate rectangle and allow fine adjustments via the keyboard, but I wasn’t having much success, so I took a path of lesser resistance.

  Packaging and Residual Weirdness
  I realized that this tool would be more useful to a broader Windows-using base of digital preservationists if they didn’t have to install Python, establish a virtual environment, and install the prerequisite dependencies. Thus, I made the effort to figure out how to wrap the entire thing up into a monolithic Windows EXE binary. It is available from the project’s Github release page (another thing I figured out for the sake of this project !).

  The binary is pretty heavy, weighing in at a bit over 50 megabytes. You might advise using compression– it IS compressed ! Before I figured out the --onefile command for pyinstaller.exe, the generated dist/ subdirectory was 150 MB. Among other things, there’s a 30 MB FORTRAN BLAS library packaged in !

  Conclusion and Future Directions
  Once I got it all working with a simple tkinter UI up front in order to select between circle and rectangle crop modes, I unleashed the tool on 60 or so scans in bulk, using the Windows forfiles command (another learning experience). I didn’t put a clock on the effort, but it felt faster. Of course, I was livid with proudness the whole time because I was using my own tool. I just wish I had thought of it sooner. But, really, with 2100+ scans under my belt, I’m just getting started– I literally have thousands more artifacts to scan for preservation.

  The tool isn’t perfect, of course. Just tonight, I threw another scan at MobyCAIRO. Just go ahead and try to find straight lines in this specimen :

  
  
  

  I eventually had to use the text left and right of center to line up against the grid with the manual keyboard adjustments. Still, I’m impressed by how these computer vision algorithms can see patterns I can’t, highlighting lines I never would have guessed at.

  I’m eager to play with OpenCV some more, particularly the video processing functions, perhaps even some GPU-accelerated versions.

  The post Developing MobyCAIRO first appeared on Breaking Eggs And Making Omelettes.

• How to Choose a GDPR Compliant Web Analytics Solution

  2 mars 2022, par Matthieu Aubry — Privacy

  Since the launch of GDPR, one big question has lingered around with uncertainty – is Google Analytics GDPR compliant ? The current GDPR enforcement trend happening across the EU is certainly shedding some light on this question.

  Starting with the Austrian Data Protection Authority’s ruling on Google Analytics and more recently, CNIL (the French Data Protection Authority) has followed suit by also ruling Google Analytics illegal to use. Organisations with EU-based web visitors are now scrambling to find a compliant solution.

  The French Data Protection Authority (CNIL) has already started delivering formal notices to websites using Google Analytics, so now is the time to act. According to CNIL, organisations have two options :

  1. Ceasing use of the Google Analytics functionality (under the current conditions) 
  2. Use a compliant web analytics tool that does not transfer data outside the EU

  Getting started 

  For organisations considering migrating to a compliant web analytics tool, I’ve outlined below the things you need to consider when weighing up compliant web analytics tools. Once you’ve made a choice, I’ve also included a step-by-step guide to migrating away from Google Analytics. This guide is useful regardless of which GDPR compliant analytics provider you choose.

  Before getting started, I recommend that you document your findings against the following considerations while reviewing GDPR compliant Google Analytics alternatives. This document can then be shared with your Data Protection Officer (DPO) to get their final recommendation.

  10 key considerations when selecting a GDPR compliant web analytics tools

  Many tools will claim to be GDPR compliant so it’s important that you do your due diligence and review tools against the following considerations. 

  1. Where does the tool store data ? 

  The rulings in France and Austria were based on the fact that Google Analytics stores data in the US, which does not have an adequate level of data protection. Your safest option is to find a tool that legally stores data in the EU.

  You should be able to find out where the data is stored in the organisation’s privacy policy. Generally, data storage information can be found under sections titled “Subprocessors” and “Third-party services”. Check out the Matomo Privacy Policy as an example. 

  If you’re unable to easily find this information or it’s unclear, reach out to the organisation for more information.

  2. Does the tool offer anonymous tracking ?

  Anonymous tracking comes with many benefits, including :

  • The ability to track visitors without a cookie consent screen. Due to the privacy-respecting aspect of cookieless tracking, you don’t need to worry about the extra steps involved with compliant cookie banners.
  • More accurate data. When visitors deny tracking cookies, you lose out on valuable data. With anonymous tracking there is no data lost as you don’t need consent to track.
  • Simplified GDPR compliance. With this enabled, there are fewer steps you need to take to get GDPR compliant and stay GDPR compliant.

  For those reasons, it may be important for you to select a tool that offers anonymous tracking functionalities. The level of anonymous tracking you require will depend on your situation but you should look out for tools that allow you to :

  • Disable fingerprinting 
  • Disable user profiles 
  • Anonymise data
  • Cookieless tracking

  If you want to read more about data anonymization, check out this guide on data anonymization in web analytics.

  3. Does the tool integrate with my existing tech stack ?

  You’ll want to ensure that a new web analytics tool will play well with other tools in your tech stack including things like your CMS (content management system), eCommerce shop, etc. You should list out all the existing tools that currently integrate with your Google Analytics and check that the same integrations can be re-created with the new tool, via integrations or APIs.

  If not, it could become costly trying to connect your existing tech stack to a new solution.

  4. Does the tool offer the same features and insights you are currently using in Google Analytics ? Or more, if necessary ? 

  Just because you are moving to a new web analytics platform, doesn’t mean you have to give up the insights, reports and features you’ve grown accustomed to with Google Analytics. Ensuring that a new platform provides the same features and reports that you value the most will result in a smoother transition away from Google Analytics.

  It’s unlikely that a new tool will have all of the same features as Google Analytics, so I’d recommend listing out and prioritising your business-critical features and reports. 

  If I had to guess, you probably set up Google Analytics years ago because it was the default option. Now is your chance to make the most of this switch from Google Analytics and find a tool that offers additional reports and features that better aligns with your business. If time permits, I’d highly recommend that you consider other features or reports that you might have been missing out on while using Google Analytics.

  Check out this comparison of Google Analytics vs Matomo to see side-by-side feature comparison.

  5. Does the tool accept Google Analytics data imports ? 

  The historical data in Google Analytics is a critical asset for many businesses. Fortunately, some tools accept Google Analytics data imports so you don’t lose all of the data you’ve generated over time.

  However, it’s important to note that any data you import from Google Analytics to a new tool needs to be compliant data. I’ll cover this more below.

  6. Does the tool provide conversion tracking exports ? 

  Do you invest in paid advertising ? If you do, then tracking the conversions from people clicking on these paid ads is critical in assessing your return on investment. Since sending IP addresses or other personal information to the US is illegal under GDPR, we can only assume that this will also apply to advertising pixel/conversion tracking (e.g., Facebook pixel, Google Ads conversion tracking, etc). 

  As an example, Matomo offers conversion tracking exports so you can get a better understanding of ad performance while meeting privacy laws and without requiring consent from users. See how it works with Matomo’s conversion tracking exports. 

  7. How will you train up your in-house team ? Or can you hire a contractor ?

  This is a common concern of many, and rightfully so. You’ll want to confirm what resources are readily available so you can hit the ground running with your new web analytics tool. If you’d prefer to train up your in-house team, check the provider’s site for training resources, videos, guides, etc.

  If you’d rather hire an external contractor, we recommend heading to LinkedIn, reaching out to your community or asking the provider if they have any recommendations for contractors.

  In addition, check that the provider offers technical support or a forum, in case you have specific questions and need help.   

  8. Does the tool offer self-hosting ? (optional)

  For organisations that want full control over their data and storage location, an on-premise web analytics tool will be the preferred option. From a GDPR perspective, this is also the easiest option for compliance.

  Keep in mind that this requires resources, regular maintenance, technical knowledge and/or technical consultants. If you’re unsure which option is best for your organisation, check out our on-premise vs cloud web analytics comparison breakdown.

  Find out more about self-hosting Matomo.

  9. Is the tool approved by the CNIL for tracking without consent ?

  This is an important step for websites with French users. This step will help narrow down your selection of tools. The CNIL offers a programme to identify web analytics solutions that can be used without tracking consent. The CNIL’s list of recommended web analytics tools can act as your starting point for solutions to review.

  While this step is specific to sites with French users, it can also be helpful for websites with visitors from any other EU country.

  Benefits of consent-free tracking

  There are many benefits of tracking without consent.
  

  For one, it simplifies GDPR compliance and reduces the chances of GDPR breaches and fines. Cookie consent screens have recently been the target for EU Data Protection Authorities because many websites are unknowingly serving cookie consent screens that do not meet GDPR requirements. 

  Yet another benefit, and quite possibly the most important is more accurate data. Even if a website displays a user-friendly, lawful consent screen, the majority of users will either ignore or reject cookie consent. Legally website owners can’t track anything unless the visitor gives consent. So not having a cookie consent screen ensures that every visit is tracked and your web analytics data is 100% accurate. 

  Lastly, many visitors have grown fatigued and frustrated with invasive cookie consent screens. Not having one on your site creates a user-friendly experience, which will likely result in longer user sessions and lower bounce rates.

  10. Does the tool offer a Data Processing Agreement (DPA) ? 

  Technically, any GDPR compliant web analytics tool should offer a DPA but for the sake of completeness, I’ve added this as a consideration. Double check that any tools you are looking at provide this legally binding document. This should be located in the Privacy Policy of the web analytics provider, if not reach out to request it.

  As an example, here’s Matomo’s Data Processing Agreement which can be found in our Privacy Policy under Subprocessors. 

  That wraps up the key considerations. When it comes to compliance, privacy and customer data, Matomo leads the way. We are looking forward to helping you achieve GDPR compliance easily. Start your free 21-day trial of Matomo now – no credit card required.

  Try Matomo Cloud

  A step-by-step guide to migrating from Google Analytics

  Once you’ve identified a tool that suits your needs and your Data Protection Officer (DPO) has approved, you’re ready to get started. Here’s a simple step-by-step guide with all the important steps for you to follow :

  1. Before getting started, you should sign or download the Data Processing Agreement (DPA) offered by your new web analytics provider.

  2. Register for the new tool and configure it for compliance. The provider should offer guides on how to configure for GDPR compliance. This will include things like giving your users an easy way to opt-out of all tracking, turning on cookieless tracking or asking users for consent and anonymizing data and IP addresses, for instance.

  3. Inform your organisation about the change. Whether your colleagues use the tool or not, it’s important that you share information about the new tool with your staff. Let them know what the tool will be used for, who will use the tool and how it complies with GDPR. 

  4. Let your DPO know that you’ve removed Google Analytics and have implemented the new tool.

  5. Update your records of processing activities to include the new tool.

  6. Update your privacy policy. You’ll need to include details about the web analytics provider, where the data is stored, what data is being collected, how long the data will be stored and why the data is being collected. The web analytics tool should readily have this information for you.

  As an example, if you decide to use Matomo as your web analytics tool, we provide a Privacy Policy template for you to use on your site and a guide on how to complete your privacy policy under GDPR with Matomo. Note that these are only applicable if you are using Matomo.

  In addition, if the tool has an opt-out feature, you will also need to put the opt-out into the privacy policy (e.g., when using cookieless tracking).

  7. Now, the exciting part. Add the tracking code to your site by following the steps provided by the web analytics tool. 

  If you’re not comfortable with this step, the provider should offer steps to do this and you can share this with your web developer.

  8. Once added, login to your tool and check to see if traffic is being tracked.

  9. If your tool does not offer Google Analytics data imports or you do not need the historical data in your new tool, go to step 11. 

  To plan for your Google Analytics data migration, you’ll first need to establish what historical data is compliant with GDPR.

  For example, you shouldn’t import any data stored beyond the retention period established in your Privacy Policy or any personally identifiable information (PII) like IP addresses that aren’t anonymised. Discuss this further with your DPO.

  10. Once you’ve established what data you can legally import, then you can begin the import. Follow the steps provided by your new web analytics solution provider.

  11. Remove Google Analytics tracking code from your site. This will stop the collection of your visitors data by Google as well as slightly increase the page load speed.

  If you still haven’t made a choice yet, try Matomo free for 21-days and see why over 1 million websites choose Matomo. 

• Buffered encoded images not saved

  26 mars 2021, par xyfix

  I have an issue with the first 12 images not being saved to file. I have attached the relevant files in this issue. I have also attached a log file to show that the first 12 images aren't written to the file that is generated. The frame rate is 24 fps and the recording is 5 sec, so there should be 120 frames written to the output file. This can be seen in the 4th column. The lines in the log files are as follows :

  


  image num [unique num from camera] [temp image num for recording seq] [time in ms]

  


  The image class is actually a simple wrapper around OpenCV's mat class with some additional members. The output file that I currently get is around 10 MB and when I open it in VLC it doesn't run for 5 seconds but more like 1 - 2 seconds but I can see whatever I have recorded. Can anyone explain to me why the files not written and the duration isn't 5 secs (minus 12 frames missing) as expected . As you can see I have tried with "av_interleaved_write_frame" but that didn't help

  


  xcodec.h

  


  #ifndef XCODEC_H
#define XCODEC_H

#include "image/Image.h"

extern "C"
{
    #include "Codec/include/libavcodec/avcodec.h"
    #include "Codec/include/libavdevice/avdevice.h"
    #include "Codec/include/libavformat/avformat.h"
    #include "Codec/include/libavutil/avutil.h"
    #include "Codec/include/libavformat/avio.h"
    #include "Codec/include/libavutil/imgutils.h"
    #include "Codec/include/libavutil/opt.h"
    #include "Codec/include/libswscale/swscale.h"
}


class XCodec
{
public:

    XCodec(const char *filename);

    ~XCodec();

    void encodeImage( const Image& image );

    void encode( AVFrame *frame, AVPacket *pkt );

    void add_stream();

    void openVideoCodec();

    void write_video_frame(const Image &image);

    void createFrame( const Image& image );

    void close();

private:

    static int s_frameCount;

    int m_timeVideo = 0;

    std::string m_filename;


    AVCodec* m_encoder = NULL;

    AVOutputFormat* m_outputFormat = NULL;

    AVFormatContext* m_formatCtx = NULL;

    AVCodecContext* m_codecCtx = NULL;

    AVStream* m_streamOut = NULL;

    AVFrame* m_frame = NULL;

    AVPacket* m_packet = NULL;

};

#endif


  


  xcodec.cpp

  


  #include "XCodec.h"&#xA;&#xA;#include <qdebug>&#xA;&#xA;&#xA;#define STREAM_DURATION   5.0&#xA;#define STREAM_FRAME_RATE 24&#xA;#define STREAM_NB_FRAMES  ((int)(STREAM_DURATION * STREAM_FRAME_RATE))&#xA;#define STREAM_PIX_FMT    AV_PIX_FMT_YUV420P /* default pix_fmt */&#xA;#define OUTPUT_CODEC AV_CODEC_ID_H264&#xA;&#xA;int XCodec::s_frameCount = 0;&#xA;&#xA;XCodec::XCodec( const char* filename ) :&#xA;    m_filename( filename ),&#xA;    m_encoder( avcodec_find_encoder( OUTPUT_CODEC ))&#xA;{&#xA;    av_log_set_level(AV_LOG_VERBOSE);&#xA;&#xA;    int ret(0);&#xA;&#xA;&#xA;    // allocate the output media context&#xA;    ret = avformat_alloc_output_context2( &amp;m_formatCtx, m_outputFormat, NULL, m_filename.c_str());&#xA;&#xA;    if (!m_formatCtx)&#xA;        return;&#xA;&#xA;    m_outputFormat = m_formatCtx->oformat;&#xA;&#xA;    // Add the video stream using H264 codec&#xA;    add_stream();&#xA;&#xA;    // Open video codec and allocate the necessary encode buffers&#xA;    if (m_streamOut)&#xA;        openVideoCodec();&#xA;&#xA;    // Print detailed information about input and output&#xA;    av_dump_format( m_formatCtx, 0, m_filename.c_str(), 1);&#xA;&#xA;    // Open the output media file, if needed&#xA;    if (!( m_outputFormat->flags &amp; AVFMT_NOFILE))&#xA;    {&#xA;        ret = avio_open( &amp;m_formatCtx->pb, m_filename.c_str(), AVIO_FLAG_WRITE);&#xA;&#xA;        if (ret &lt; 0)&#xA;        {&#xA;            char error[255];&#xA;            ret = av_strerror( ret, error, 255);&#xA;            fprintf(stderr, "Could not open &#x27;%s&#x27;: %s\n", m_filename.c_str(), error);&#xA;            return ;&#xA;        }&#xA;    }&#xA;    else&#xA;    {&#xA;        return;&#xA;    }&#xA;&#xA;    // Write media header&#xA;    ret = avformat_write_header( m_formatCtx, NULL );&#xA;&#xA;    if (ret &lt; 0)&#xA;    {&#xA;        char error[255];&#xA;        av_strerror(ret, error, 255);&#xA;        fprintf(stderr, "Error occurred when opening output file: %s\n", error);&#xA;        return;&#xA;    }&#xA;&#xA;    if ( m_frame )&#xA;           m_frame->pts = 0;&#xA;}&#xA;&#xA;&#xA;&#xA;XCodec::~XCodec()&#xA;{}&#xA;&#xA;/* Add an output stream. */&#xA;void XCodec::add_stream()&#xA;{&#xA;    AVCodecID codecId = OUTPUT_CODEC;&#xA;&#xA;    if (!( m_encoder ))&#xA;    {&#xA;        fprintf(stderr, "Could not find encoder for &#x27;%s&#x27;\n",&#xA;            avcodec_get_name(codecId));&#xA;        return;&#xA;    }&#xA;&#xA;    // Get the stream for codec&#xA;    m_streamOut = avformat_new_stream(m_formatCtx, m_encoder);&#xA;&#xA;    if (!m_streamOut) {&#xA;        fprintf(stderr, "Could not allocate stream\n");&#xA;        return;&#xA;    }&#xA;&#xA;    m_streamOut->id = m_formatCtx->nb_streams - 1;&#xA;&#xA;    m_codecCtx = avcodec_alloc_context3( m_encoder);&#xA;&#xA;    switch (( m_encoder)->type)&#xA;    {&#xA;    case AVMEDIA_TYPE_VIDEO:&#xA;        m_streamOut->codecpar->codec_id = codecId;&#xA;        m_streamOut->codecpar->codec_type = AVMEDIA_TYPE_VIDEO;&#xA;        m_streamOut->codecpar->bit_rate = 400000;&#xA;        m_streamOut->codecpar->width = 800;&#xA;        m_streamOut->codecpar->height = 640;&#xA;        m_streamOut->codecpar->format = STREAM_PIX_FMT;&#xA;        m_streamOut->time_base = { 1, STREAM_FRAME_RATE };&#xA;&#xA;        avcodec_parameters_to_context( m_codecCtx, m_streamOut->codecpar);&#xA;&#xA;        m_codecCtx->gop_size = 12; /* emit one intra frame every twelve frames at most */&#xA;        m_codecCtx->max_b_frames = 1;&#xA;        m_codecCtx->time_base = { 1, STREAM_FRAME_RATE };&#xA;        m_codecCtx->framerate = { STREAM_FRAME_RATE, 1 };&#xA;        m_codecCtx->pix_fmt = STREAM_PIX_FMT;&#xA;        m_codecCtx->profile = FF_PROFILE_H264_HIGH;&#xA;&#xA;        break;&#xA;&#xA;    default:&#xA;        break;&#xA;    }&#xA;&#xA;    if (m_streamOut->codecpar->codec_id == OUTPUT_CODEC)&#xA;    {&#xA;      av_opt_set( m_codecCtx, "preset", "ultrafast", 0 );&#xA;    }&#xA;&#xA;/&#xA;//    /* Some formats want stream headers to be separate. */&#xA;    if (m_formatCtx->oformat->flags &amp; AVFMT_GLOBALHEADER)&#xA;            m_codecCtx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;&#xA;&#xA;&#xA;    int ret = avcodec_parameters_from_context( m_streamOut->codecpar, m_codecCtx );&#xA;&#xA;    if (ret &lt; 0)&#xA;    {&#xA;        char error[255];&#xA;        av_strerror(ret, error, 255);&#xA;        fprintf(stderr, "avcodec_parameters_from_context returned (%d) - %s", ret, error);&#xA;        return;&#xA;    }&#xA;}&#xA;&#xA;&#xA;void XCodec::openVideoCodec()&#xA;{&#xA;    int ret;&#xA;&#xA;    /* open the codec */&#xA;    ret = avcodec_open2(m_codecCtx, m_encoder, NULL);&#xA;&#xA;    if (ret &lt; 0)&#xA;    {&#xA;        char error[255];&#xA;        av_strerror(ret, error, 255);&#xA;        fprintf(stderr, "Could not open video codec: %s\n", error);&#xA;        return;&#xA;    }&#xA;&#xA;    /* allocate and init a re-usable frame */&#xA;//    m_frame = av_frame_alloc();&#xA;&#xA;}&#xA;&#xA;&#xA;void XCodec::encodeImage(const Image &amp;image)&#xA;{&#xA;    // Compute video time from last added video frame&#xA;    m_timeVideo = image.timeStamp(); //(double)m_frame->pts) * av_q2d(m_streamOut->time_base);&#xA;&#xA;    // Stop media if enough time&#xA;    if (!m_streamOut /*|| m_timeVideo >= STREAM_DURATION*/)&#xA;       return;&#xA;&#xA;&#xA;    // Add a video frame&#xA;    write_video_frame( image );&#xA;&#xA;}&#xA;&#xA;&#xA;void XCodec::write_video_frame( const Image&amp; image )&#xA;{&#xA;    int ret;&#xA;&#xA;qDebug() &lt;&lt; "image num " &lt;&lt; image.uniqueImageNumber() &lt;&lt; " " &lt;&lt; s_frameCount;&#xA;&#xA;    if ( s_frameCount >= STREAM_NB_FRAMES)&#xA;    {&#xA;        /* No more frames to compress. The codec has a latency of a few&#xA;         * frames if using B-frames, so we get the last frames by&#xA;         * passing the same picture again. */&#xA;        int p( 0 ) ;&#xA;    }&#xA;    else&#xA;    {&#xA;         createFrame( image );&#xA;    }&#xA;&#xA;    // Increase frame pts according to time base&#xA;//    m_frame->pts &#x2B;= av_rescale_q(1, m_codecCtx->time_base, m_streamOut->time_base);&#xA;    m_frame->pts = int64_t( image.timeStamp()) ;&#xA;&#xA;&#xA;    if (m_formatCtx->oformat->flags &amp; 0x0020 )&#xA;    {&#xA;        /* Raw video case - directly store the picture in the packet */&#xA;        AVPacket pkt;&#xA;        av_init_packet(&amp;pkt);&#xA;&#xA;        pkt.flags |= AV_PKT_FLAG_KEY;&#xA;        pkt.stream_index = m_streamOut->index;&#xA;        pkt.data = m_frame->data[0];&#xA;        pkt.size = sizeof(AVPicture);&#xA;&#xA;//        ret = av_interleaved_write_frame(m_formatCtx, &amp;pkt);&#xA;        ret = av_write_frame( m_formatCtx, &amp;pkt );&#xA;    }&#xA;    else&#xA;    {&#xA;        AVPacket pkt;&#xA;        av_init_packet(&amp;pkt);&#xA;&#xA;        /* encode the image */&#xA;        ret = avcodec_send_frame(m_codecCtx, m_frame);&#xA;&#xA;        if (ret &lt; 0)&#xA;        {&#xA;            char error[255];&#xA;            av_strerror(ret, error, 255);&#xA;            fprintf(stderr, "Error encoding video frame: %s\n", error);&#xA;            return;&#xA;        }&#xA;&#xA;        /* If size is zero, it means the image was buffered. */&#xA;        ret = avcodec_receive_packet(m_codecCtx, &amp;pkt);&#xA;&#xA;        if( !ret &amp;&amp; pkt.size)&#xA;        {&#xA;qDebug() &lt;&lt; "write frame " &lt;&lt; m_frame->display_picture_number;&#xA;            pkt.stream_index = m_streamOut->index;&#xA;&#xA;            /* Write the compressed frame to the media file. */&#xA;//            ret = av_interleaved_write_frame(m_formatCtx, &amp;pkt);&#xA;            ret = av_write_frame( m_formatCtx, &amp;pkt );&#xA;        }&#xA;        else&#xA;        {&#xA;            ret = 0;&#xA;        }&#xA;    }&#xA;&#xA;    if (ret != 0)&#xA;    {&#xA;        char error[255];&#xA;        av_strerror(ret, error, 255);&#xA;        fprintf(stderr, "Error while writing video frame: %s\n", error);&#xA;        return;&#xA;    }&#xA;&#xA;    s_frameCount&#x2B;&#x2B;;&#xA;}&#xA;&#xA;&#xA;void XCodec::createFrame( const Image&amp; image /*, AVFrame *m_frame, int frame_index, int width, int height*/)&#xA;{&#xA;    /**&#xA;     * \note allocate frame&#xA;     */&#xA;    m_frame = av_frame_alloc();&#xA;    int ret = av_frame_make_writable( m_frame );&#xA;&#xA;    m_frame->format = STREAM_PIX_FMT;&#xA;    m_frame->width = image.width();&#xA;    m_frame->height = image.height();&#xA;//    m_frame->pict_type = AV_PICTURE_TYPE_I;&#xA;    m_frame->display_picture_number = image.uniqueImageNumber();&#xA;&#xA;    ret = av_image_alloc(m_frame->data, m_frame->linesize, m_frame->width,  m_frame->height, STREAM_PIX_FMT, 1);&#xA;&#xA;    if (ret &lt; 0)&#xA;    {&#xA;        return;&#xA;    }&#xA;&#xA;    struct SwsContext* sws_ctx = sws_getContext((int)image.width(), (int)image.height(), AV_PIX_FMT_RGB24,&#xA;                                                (int)image.width(), (int)image.height(), STREAM_PIX_FMT, 0, NULL, NULL, NULL);&#xA;&#xA;    const uint8_t* rgbData[1] = { (uint8_t* )image.getData() };&#xA;    int rgbLineSize[1] = { 3 * image.width() };&#xA;&#xA;    sws_scale(sws_ctx, rgbData, rgbLineSize, 0, image.height(), m_frame->data, m_frame->linesize);&#xA;&#xA;//cv::Mat yuv420p( m_frame->height &#x2B; m_frame->height/2, m_frame->width, CV_8UC1, m_frame->data[0]);&#xA;//cv::Mat cvmIm;&#xA;//cv::cvtColor(yuv420p,cvmIm,CV_YUV420p2BGR);&#xA;//std::ostringstream ss;&#xA;//ss &lt;&lt; "c:\\tmp\\YUVoriginal_" &lt;&lt; image.uniqueImageNumber() &lt;&lt; ".png";&#xA;//cv::imwrite( ss.str().c_str(), cvmIm);&#xA;}&#xA;&#xA;&#xA;void XCodec::close()&#xA;{&#xA;    /* reset the framecount */&#xA;    s_frameCount = 0 ;&#xA;&#xA;    int ret( 0 );&#xA;&#xA;    /* flush the encoder */&#xA;    while( ret >= 0 )&#xA;        ret = avcodec_send_frame(m_codecCtx, NULL);&#xA;&#xA;    // Write media trailer&#xA;    if( m_formatCtx )&#xA;        ret = av_write_trailer( m_formatCtx );&#xA;&#xA;    /* Close each codec. */&#xA;    if ( m_streamOut )&#xA;    {&#xA;        if( m_frame )&#xA;        {&#xA;            av_free( m_frame->data[0]);&#xA;            av_frame_free( &amp;m_frame );&#xA;        }&#xA;&#xA;        if( m_packet )&#xA;            av_packet_free( &amp;m_packet );&#xA;    }&#xA;&#xA;    if (!( m_outputFormat->flags &amp; AVFMT_NOFILE))&#xA;        /* Close the output file. */&#xA;        ret = avio_close( m_formatCtx->pb);&#xA;&#xA;&#xA;    /* free the stream */&#xA;    avformat_free_context( m_formatCtx );&#xA;&#xA;    fflush( stdout );&#xA;}&#xA;</qdebug>

  


  image.h

  


  #ifndef IMAGE_H
#define IMAGE_H

#include  

class Image 
{
public:

    Image();

    Image( const cv::Mat& mat );

    Image(const Image& other) = default;

    Image(Image&& other) = default;

    ~Image();


    inline const cv::Mat& matrix() const{ return m_matrix; }

    inline const int uniqueImageNumber() const{ return m_uniqueId; }

    inline const int timeStamp() const { return m_timeStamp; }

    inline const int width() const { return m_matrix.cols(); }
    
    inline const int height() const { return m_matrix.rows(); }

private:

    cv::Mat   m_matrix;

    int       m_timeStamp;

    int       m_uniqueId;

};

#endif


  


  logtxt

  


   image num  1725   0   0
 image num  1727   1   40
 image num  1729   2   84
 image num  1730   3   126
 image num  1732   4   169
 image num  1734   5   211
 image num  1736   6   259
 image num  1738   7   297
 image num  1740   8   340
 image num  1742   9   383
 image num  1744   10   425
 image num  1746   11   467
 image num  1748   12   511
 image num  1750   13   553
 write frame  1750
 image num  1752   14   600
 write frame  1752
 image num  1753   15   637
 write frame  1753
 image num  1755   16   680
 write frame  1755
 image num  1757   17   723
 write frame  1757
 image num  1759   18   766
 write frame  1759
 image num  1761   19   808
 write frame  1761
 image num  1763   20   854
 write frame  1763
 image num  1765   21   893
 write frame  1765
 image num  1767   22   937
 write frame  1767
 image num  1769   23   979
 write frame  1769
 image num  1770   24   1022
 write frame  1770
 image num  1772   25   1064
 write frame  1772
 image num  1774   26   1108
 write frame  1774
 image num  1776   27   1150
 write frame  1776
 image num  1778   28   1192
 write frame  1778
 image num  1780   29   1235
 write frame  1780
 image num  1782   30   1277
 write frame  1782
 image num  1784   31   1320
 write frame  1784
 image num  1786   32   1362
 write frame  1786
 image num  1787   33   1405
 write frame  1787
 image num  1789   34   1450
 write frame  1789
 image num  1791   35   1493
 write frame  1791
 image num  1793   36   1536
 write frame  1793
 image num  1795   37   1578
 write frame  1795
 image num  1797   38   1621
 write frame  1797
 image num  1799   39   1663
 write frame  1799
 image num  1801   40   1709
 write frame  1801
 image num  1803   41   1748
 write frame  1803
 image num  1805   42   1791
 write frame  1805
 image num  1807   43   1833
 write frame  1807
 image num  1808   44   1876
 write frame  1808
 image num  1810   45   1920
 write frame  1810
 image num  1812   46   1962
 write frame  1812
 image num  1814   47   2004
 write frame  1814
 image num  1816   48   2048
 write frame  1816
 image num  1818   49   2092
 write frame  1818
 image num  1820   50   2133
 write frame  1820
 image num  1822   51   2175
 write frame  1822
 image num  1824   52   2221
 write frame  1824
 image num  1826   53   2277
 write frame  1826
 image num  1828   54   2319
 write frame  1828
 image num  1830   55   2361
 write frame  1830
 image num  1832   56   2405
 write frame  1832
 image num  1833   57   2447
 write frame  1833
 image num  1835   58   2491
 write frame  1835
 image num  1837   59   2533
 write frame  1837
 image num  1839   60   2576
 write frame  1839
 image num  1841   61   2619
 write frame  1841
 image num  1843   62   2662
 write frame  1843
 image num  1845   63   2704
 write frame  1845
 image num  1847   64   2746
 write frame  1847
 image num  1849   65   2789
 write frame  1849
 image num  1851   66   2831
 write frame  1851
 image num  1852   67   2874
 write frame  1852
 image num  1854   68   2917
 write frame  1854
 image num  1856   69   2959
 write frame  1856
 image num  1858   70   3003
 write frame  1858
 image num  1860   71   3045
 write frame  1860
 image num  1862   72   3088
 write frame  1862
 image num  1864   73   3130
 write frame  1864
 image num  1866   74   3173
 write frame  1866
 image num  1868   75   3215
 write frame  1868
 image num  1870   76   3257
 write frame  1870
 image num  1872   77   3306
 write frame  1872
 image num  1873   78   3347
 write frame  1873
 image num  1875   79   3389
 write frame  1875
 image num  1877   80   3433
 write frame  1877
 image num  1879   81   3475
 write frame  1879
 image num  1883   82   3562
 write frame  1883
 image num  1885   83   3603
 write frame  1885
 image num  1887   84   3660
 write frame  1887
 image num  1889   85   3704
 write frame  1889
 image num  1891   86   3747
 write frame  1891
 image num  1893   87   3789
 write frame  1893
 image num  1895   88   3832
 write frame  1895
 image num  1897   89   3874
 write frame  1897
 image num  1899   90   3917
 write frame  1899
 image num  1900   91   3959
 write frame  1900
 image num  1902   92   4001
 write frame  1902
 image num  1904   93   4044
 write frame  1904
 image num  1906   94   4086
 write frame  1906
 image num  1908   95   4130
 write frame  1908
 image num  1910   96   4174
 write frame  1910
 image num  1912   97   4216
 write frame  1912
 image num  1914   98   4257
 write frame  1914
 image num  1915   99   4303
 write frame  1915
 image num  1918   100   4344
 write frame  1918
 image num  1919   101   4387
 write frame  1919
 image num  1922   102   4451
 write frame  1922
 image num  1924   103   4494
 write frame  1924
 image num  1926   104   4541
 write frame  1926
 image num  1927   105   4588
 write frame  1927
 image num  1931   106   4665
 write frame  1931
 image num  1933   107   4707
 write frame  1933
 image num  1935   108   4750
 write frame  1935
 image num  1937   109   4794
 write frame  1937
 image num  1939   110   4836
 write frame  1939
 image num  1941   111   4879
 write frame  1941
 image num  1943   112   4922
 write frame  1943
 image num  1945   113   4965
 write frame  1945
 image num  1947   114   5007
 write frame  1947
 image num  1948   115   5050
 write frame  1948
 image num  1950   116   5093
 write frame  1950
 image num  1952   117   5136
 write frame  1952
 image num  1954   118   5178
 write frame  1954
 image num  1956   119   5221
 write frame  1956
 MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
0, 8-bit
Not writing 'clli' atom. No content light level info.
Not writing 'mdcv' atom. Missing mastering metadata.
 2 seeks, 41 writeouts