trsent
03-25-2006, 12:41 PM
A very interesting post from the MEARS board, I thought I would share it with us photo-matching experts for some thoughts.
http://www.network54.com/Forum/426247/message/1143307531/Imagery+Analysis-+Often+Overlooked+Aspects+and+Techniques
----------------------------------------------------------------------
Imagery Analysis: Often Overlooked Aspects and Techniques
March 25 2006 at 11:25 AM Dave Grob (davegrob1@aol.com) Dave Grob (Login davegrob1 (http://www.network54.com/Profile/davegrob1))Recently I was asked to do imagery analysis on a historically significant home flannel jersey. Please keep in mind this is not the same as offering an opinion on the item itself, but rather an important sub-set of the overall process. Since I was not able to travel to see the jersey, and it did not seem prudent to send me this six figure+ jersey in the mail, I asked that full color copies of the jersey be sent along with the images from the historical event.
If you have never done this, by that I mean obtaining a full color copy of a uniform, you may consider doing it as it will serve as a training aid to assist you in doing imagery analysis. All this can be done very easily by taking the jersey or bat to say the local Kinko’s and making 11 x17 full color images. This does a couple of things. It gives you the ability work with an item in flat manner that makes it easier to record measurements. It also provides you with a means to “travel” with your item if you wish to have someone provide some initial feedback. The final thing this does for you is permit you to retain valuable information about the jersey should you decide to sell it or trade it at a later date.
You will note if not referred to this practice as “photo matching” as it a phase I detest. Imagery analysis involves not just a visual comparison between a jersey and an image, but analyzing, measuring, and evaluating the findings in order to form an opinion.
Imagery analysis starts with the image of the actual item and not the jersey in question. Many people will start with the jersey and look for items that “match” in the picture. This is both backwards and will often lead people to stop the process when they find what they feel they are looking for.
In this case, the images I looked at where without an object of know size that would permit me to add scale to the image. This does not mean that these pictures have no empirical value. By this I mean the imagery analyst is then tasked to establish a sense of proportionality that can be used to serve as an objective basis for comparison. In this instance, the jersey in the picture was a 1951 NY Yankees home jersey.
What I did was take a measurement of the NY crest along the vertical axis from the game photos. This measurement came out to be 15mm. I then took a measurement of the left sleeve. This was a set-in as opposed to a Raglan sleeve so that the start point for the shoulder measurement was easier to determine. This measurement came out to be 18mm. This means, that for the jersey that was worn by the player on this historical event, the NY crest was approximately 83.3% of the size of the sleeve length. This is a know fact that can now be used to evaluate the jersey in question. I next repeated the same process on the full color/to scale color copies of the shirt. I found that the NY crest on the actual jersey came out to 136mm and the sleeve length came out to be 163mm. This means that the crest on the jersey in question was approximately 83.4% of the size of the sleeve length. What this means is that that these two characteristics on both jerseys are proportionally the same. While not conclusive in its own right, it does establish a series of facts about both the jersey in question and the one worn in the game that is not guess work.
The other thing I did in the course of this imagery analysis was to look at the NY crest with respect to its relative placement and alignment with respect to the pinstripes. Another commonly overlooked aspect of this process is that people will only focus on the 14 external points of the NY crest to see where they align with the pinstripes. By 14 points, I am not referring those that where offered by President Woodrow Wilson in his address to Congress on 8 January 1918 as part of the Peace Plan to conclude the First World War. Rather, I am referring to external points of the NY crest where they change direction. Back to the topic at hand. There are two other points that have to be analyzed and assessed as well. If you look at a NY Yankees home pinstriped jersey, you will notice that the pinstripe pattern also appears or not in a space between the upper portions of the “N” and the “Y”. This is important to note and asses as it is indicative of the vertical placement of the crest on the jersey itself. This goes back to my suggestion of making a full a color copy of a jersey and practicing some of this for yourself.
1. Make two copies of a jersey in this manner.
2. On one of the copies, cut out the NY crest.
3. Place the cut out over the one on the full jersey.
4. Slide the cut out crest up or down keeping the outer 14 points matched to the alignment of the pinstripes.
What you will notice is that although the points of reference with respect to the pinstripes can remain the same for the eternal 14 points, you will see that the position and or presence of the pinstripe within the space between the upper portions of the “N” and the “Y” change.
This aspect of imagery analysis is important to understand and master as it is frequently not possible to establish horizontal alignment/placement of a crest with respect to button placement because of either player body position or buttons being obscured by things like a players arms as he swings through with a bat in his hands. In the case a of pitcher, often these details are obscured by his hands and glove being brought up to the chest.
These same things can be applied to any number of teams who wore pinstriped uniforms and is not unique to the NY Yankees. Another problem I have seen with the “photo matchers” deals with matching of the pinstripes in the shoulder area for set-in sleeve jerseys. A set-in sleeve is one that is sewn to outer part of the shoulder area. A Raglan sleeve is one that is part of the shoulder construction and runs all the way to the neck line.
For set-in sleeves, pay attention to how the jersey in the picture is constructed. By this I mean that is the front half of the jersey fabric sewn over the rear half or is the rear half sewn over the front half. This subtle detail can provide insights as to who manufactured the jersey. While the pinstripes might “photo match”, you may be dealing with a salesman sample from another manufacturer as these are frequently made up for a teams star player of the day.
One of my long term goals is to establish some fundamental principles and guidelines for imagery analysis that are applicable for any number of products, not limited to baseball jerseys. The physical and intellectual process is the same for just about anything, you just need to develop the right metrics.
Dave Grob
http://www.network54.com/Forum/426247/message/1143307531/Imagery+Analysis-+Often+Overlooked+Aspects+and+Techniques
----------------------------------------------------------------------
Imagery Analysis: Often Overlooked Aspects and Techniques
March 25 2006 at 11:25 AM Dave Grob (davegrob1@aol.com) Dave Grob (Login davegrob1 (http://www.network54.com/Profile/davegrob1))Recently I was asked to do imagery analysis on a historically significant home flannel jersey. Please keep in mind this is not the same as offering an opinion on the item itself, but rather an important sub-set of the overall process. Since I was not able to travel to see the jersey, and it did not seem prudent to send me this six figure+ jersey in the mail, I asked that full color copies of the jersey be sent along with the images from the historical event.
If you have never done this, by that I mean obtaining a full color copy of a uniform, you may consider doing it as it will serve as a training aid to assist you in doing imagery analysis. All this can be done very easily by taking the jersey or bat to say the local Kinko’s and making 11 x17 full color images. This does a couple of things. It gives you the ability work with an item in flat manner that makes it easier to record measurements. It also provides you with a means to “travel” with your item if you wish to have someone provide some initial feedback. The final thing this does for you is permit you to retain valuable information about the jersey should you decide to sell it or trade it at a later date.
You will note if not referred to this practice as “photo matching” as it a phase I detest. Imagery analysis involves not just a visual comparison between a jersey and an image, but analyzing, measuring, and evaluating the findings in order to form an opinion.
Imagery analysis starts with the image of the actual item and not the jersey in question. Many people will start with the jersey and look for items that “match” in the picture. This is both backwards and will often lead people to stop the process when they find what they feel they are looking for.
In this case, the images I looked at where without an object of know size that would permit me to add scale to the image. This does not mean that these pictures have no empirical value. By this I mean the imagery analyst is then tasked to establish a sense of proportionality that can be used to serve as an objective basis for comparison. In this instance, the jersey in the picture was a 1951 NY Yankees home jersey.
What I did was take a measurement of the NY crest along the vertical axis from the game photos. This measurement came out to be 15mm. I then took a measurement of the left sleeve. This was a set-in as opposed to a Raglan sleeve so that the start point for the shoulder measurement was easier to determine. This measurement came out to be 18mm. This means, that for the jersey that was worn by the player on this historical event, the NY crest was approximately 83.3% of the size of the sleeve length. This is a know fact that can now be used to evaluate the jersey in question. I next repeated the same process on the full color/to scale color copies of the shirt. I found that the NY crest on the actual jersey came out to 136mm and the sleeve length came out to be 163mm. This means that the crest on the jersey in question was approximately 83.4% of the size of the sleeve length. What this means is that that these two characteristics on both jerseys are proportionally the same. While not conclusive in its own right, it does establish a series of facts about both the jersey in question and the one worn in the game that is not guess work.
The other thing I did in the course of this imagery analysis was to look at the NY crest with respect to its relative placement and alignment with respect to the pinstripes. Another commonly overlooked aspect of this process is that people will only focus on the 14 external points of the NY crest to see where they align with the pinstripes. By 14 points, I am not referring those that where offered by President Woodrow Wilson in his address to Congress on 8 January 1918 as part of the Peace Plan to conclude the First World War. Rather, I am referring to external points of the NY crest where they change direction. Back to the topic at hand. There are two other points that have to be analyzed and assessed as well. If you look at a NY Yankees home pinstriped jersey, you will notice that the pinstripe pattern also appears or not in a space between the upper portions of the “N” and the “Y”. This is important to note and asses as it is indicative of the vertical placement of the crest on the jersey itself. This goes back to my suggestion of making a full a color copy of a jersey and practicing some of this for yourself.
1. Make two copies of a jersey in this manner.
2. On one of the copies, cut out the NY crest.
3. Place the cut out over the one on the full jersey.
4. Slide the cut out crest up or down keeping the outer 14 points matched to the alignment of the pinstripes.
What you will notice is that although the points of reference with respect to the pinstripes can remain the same for the eternal 14 points, you will see that the position and or presence of the pinstripe within the space between the upper portions of the “N” and the “Y” change.
This aspect of imagery analysis is important to understand and master as it is frequently not possible to establish horizontal alignment/placement of a crest with respect to button placement because of either player body position or buttons being obscured by things like a players arms as he swings through with a bat in his hands. In the case a of pitcher, often these details are obscured by his hands and glove being brought up to the chest.
These same things can be applied to any number of teams who wore pinstriped uniforms and is not unique to the NY Yankees. Another problem I have seen with the “photo matchers” deals with matching of the pinstripes in the shoulder area for set-in sleeve jerseys. A set-in sleeve is one that is sewn to outer part of the shoulder area. A Raglan sleeve is one that is part of the shoulder construction and runs all the way to the neck line.
For set-in sleeves, pay attention to how the jersey in the picture is constructed. By this I mean that is the front half of the jersey fabric sewn over the rear half or is the rear half sewn over the front half. This subtle detail can provide insights as to who manufactured the jersey. While the pinstripes might “photo match”, you may be dealing with a salesman sample from another manufacturer as these are frequently made up for a teams star player of the day.
One of my long term goals is to establish some fundamental principles and guidelines for imagery analysis that are applicable for any number of products, not limited to baseball jerseys. The physical and intellectual process is the same for just about anything, you just need to develop the right metrics.
Dave Grob