In this editorial, I would like to give some information on the reviewing and the publication process of the Journal of Mathematical Imaging and Vision (JMIV).

The quality of every journal depends equally on the quality of its submissions and the quality of the reviewing process. Serving as reviewer for our scientific community may look little attractive at first glance: it requires some extra time on top of the daily commitments, and it seems to give not much glory since reviewers are supposed to remain anonymous. A closer look, however, gives a different impression.

Our peer reviewing system is a system of give and take: writing manuscripts and reviewing them are two sides of the same coin. Authors benefit from the expertise of their colleagues who help them spot errors and give suggestions for improved writing. Reviewers may find it interesting to learn more about the latest developments in the field. Giving fair and constructive criticism and judging scientific merits is also something that in particular younger researchers can learn from their activities as a reviewer. A scientific community, where a manuscript is inspected by three reviewers, can only function in a fair way if one is willing to review on average three manuscripts for every submitted paper. Thus, volunteering as a reviewer for a specific journal is also a matter of fairness and of giving something back.

It is misleading to believe that thorough work as a reviewer is not appreciated very much and remains unknown to others. Needless to say, authors do appreciate good reviews. While they do not know the name of their reviewers, the coordinating editor and the editor-in-chief do. Like many other journals, JMIV has an electronic system that allows them to grade the quality of a review with mark between 1 and 100 %. Together with additional statistical data such as the average time for each review and the number of accepted or declined invitations to review a manuscript, this gives fairly comprehensive information on the performance of each reviewer. This information is visible to all members of the editorial board when they invite reviewers for a specific manuscript. These experts belong to the leading researchers in their field, and they are the ones who write recommendation letters that may help boost the career of their colleagues. Also the invitation to serve in the editorial board of JMIV does not only reflect the scientific merits of a top researcher, but also the dedication and thoroughness as a reviewer.

In order to increase the awareness of the importance of thorough reviewing, it was decided that every spring we publish the names of all reviewers who have supported JMIV in the previous year. The present issue contains such an appreciation of our reviewers. Please have a look at their names. Many of them are very well-known researchers who allocate part of the precious time to support our journal and the scientific community. These are the experts who give you valuable feedback on your submitted manuscript. It is my pleasure to thank them all for their hard work.

Two questions I am asked very often are concerned with the average publication time of JMIV and its acceptance rate. Before discussing these numbers, let us have a look at the total workflow first. Every regular manuscript is first sent to the editor-in-chief. His task is to get a first impression and judge whether the manuscript falls within the scope of JMIV. There is a surprisingly large number of submissions that do not offer the substantial mathematical depth that is required for our journal. Others are obviously not of sufficient scientific quality, or their English is below the minimal acceptable level for a journal publication. These manuscripts are rejected right away, usually within 1 week after submission. The remaining submissions are sent to a coordinating editor whose scientific expertise covers the topic of the manuscript. Also this person has the option to reject a manuscipt if it appears to have no chance in the reviewing process. If this is not the case, the coordinating editor invites potential reviewers who usually have 28 days to review a manuscipt. This process can take some additional time, since some of those people do not respond, others do not agree, and those who agree do not always provide their review in time. Once all reviews are in, the coordinating editor gives a recommendation. Then the manuscript is sent to the editor-in-chief who reads all reviews along with the recommendation of the coordinating editor and makes a final decision. If revisions are required, authors resubmit their paper, and the review cycle starts again. Usually, minor revisions are handled by the coordinating editor, while major revisions are sent back to the reviewers. Thus, numerous steps are involved to secure the quality of each published manuscript.

While this workflow is similar for most journals, how is the specific situation for JMIV? In 2013, we received 262 regular manuscripts. Of them, 150 (57.3 %) have been rejected directly by the editor-in-chief or the coordinating editor, 65 (24.8 %) have been rejected after reviewing, and 47 (17.9 %) have been accepted. An acceptance rate of 17.9 % is on par with the selectivity and the prestige of the very best computer vision conferences. On the other hand, if one sees it as a two-stage process, then each manuscript that undergoes reviewing has a fair chance of 42 % of being accepted. Authors of manuscripts that are rejected without reviewing do not lose much time and can improve their manuscripts before submitting them elsewhere.

At JMIV, we have worked very hard to speed up the reviewing process. Editors and reviewers who are behind their schedule are reminded on a regular basis. As a result, in 2014 the median time from submission to acceptance of a manuscript could be reduced to as little as 7 months. Please note that this involves on average two to three review rounds. Once a manuscipt is accepted for publication, it will appear as Online First paper within about 2 weeks. Thus, with a median publication period of seven and a half months, JMIV is even faster than leading major conferences on image processing and computer vision. Moreover, authors benefit from several review cycles where they receive valuable feedback that helps them publish a definitive version of their work in the highest possible quality. In this context, it should also be mentioned that JMIV does not impose any limit on the number of pages, and that its impact factor has increased from 1.391 to 2.330 within the last 2 years.

After each manuscript has appeared as Online First version, it is also scheduled for a printed issue, where it is assigned its final volume and page numbers. Naturally, this process takes some additional time, since there are several earlier papers in the pipeline. However, we have also taken measures to reduce the backlog time substantially: if our stack of accepted manuscripts has grown more than necessary for a stable planning, we publish larger issues with more manuscripts. In this way, we can implement a backlog that does not exceed half a year. This is lower than most other journals.

Please have a look at the current issue to convince yourself that these goals are really achieved in practice. Five out of eight papers have been accepted within 6 months, and the typical time from acceptance to publication as Online First paper is about 2  weeks. Moreover, all papers in this issue have appeared online about half a year ago.

If you are an author, let me assure you that also in the future, we will continue our efforts to maintain JMIV as a top journal for publishing your best manuscripts on mathematical image analysis in a smooth and efficient way. For you as a reader, I hope you will appreciate the numerous efforts that we have taken to provide you with publications of highest quality. Enjoy.